MVSSA Conference 22 23 June 2023 Proceedings

THE
MINE VENTILATION SOCIETY

OF SOUTH AFRICA
2023 Annual Conference

“Ventilation Engineers Going into the future using diverse
Technologies and Innovations”.
22 and 23 June 2023

Emperors Palace
Ekurhuleni
MVSSA Annual Conference 2023: “Ventilation Engineers going into the future using diverse technologies and innovations”
© 2023 The Mine Ventilation Society of South Africa
Published by:
The Mine Ventilation Society of South Africa

Johannesburg
© The Mine Ventilation Society of South Africa, 2023
ISBN 978-0-6397-9604-8
The inclusion of descriptions, or discussions of, particular products in the Conference Proceedings does not im-
ply that The Mine Ventilation Society of South Africa [MVSSA] endorses any of these
products.
ii
Foreword
Another exciting year to physically host and present the Proceedings of the 2023 Mine Ventilation Society of
South Africa Conference.
The values and objectives on which the Mine Ventilation Society of South Africa (MVSSA) was built are still
more relevant in promoting and advancing the science and practices of ventilation engineering, with reference
to mine ventilation, air quality, and related challenges. As mining continues in South Africa old and new chal-
lenges emerge ranging from aging infrastructures and knowledgeable professionals, to deepening mineral re-
sources, with opportunities for new development in knowledge and technology.
The MVSSA Annual Conference aims to create a platform where multi-stakeholder in occupational health space
and engineering, academics and original equipment manufacturers meet and discuss mine ventilation improve-
ments, risks and hazards management, training and education, and any other related issues. This is with an
endeavour to create a conducive environment for employees in a cost-effective manner; whilst considering and
adhering to the legislative framework.
The MVSSA conference is being held for two days under the theme “Ventilation Engineers Going into the
future using diverse Technologies and Innovations”. The keynote address will focus on the works of the Medical
Bureau for Occupational Disease (MBOD) and the newly introduced risk determination methods and implica-
tions for the Ventilation Engineering discipline, i.e., enhance control measures.
The topics that will be presented at the conference are divided into five (5) categories: -
Mine Ventilation engineering - where some of the topics will share knowledge on Planning and designing an
efficient mine network; tailoring plant engineering to fast-tracking and construction refrigeration system; and
preparing a mine ventilation network to support expansion on the underground workings.
Environmental Impact and Occupational Hygiene - sharing of the studies conducted such as: Investigating the
business case for implementing EBM; Ergonomics risk factors affecting women; a realistic carbon offset for
the South African Mining Industry; The role of artificial intelligence in enhancing Mine Ventilation, Emergency
Preparedness, and Occupational Hygiene; and an additive for Silica scrubbing.
Risk Management - practices are shared such as: the application of Continuous Real-Time Monitoring (CRTM)
in Noise Risk Management; establishing a process for the setting of real-time detection system alarms; real-
time detection of nitrogen dioxide; and issues, concerns, setting alarms, and interpreting data will be presented.
Emergency preparedness and audits learnings will be shared on: remote safety audits – drawbacks and hidden
treasures; simulation of conveyor belt and vehicle fires in South African Coal Mines; best practices and lessons
learnt in emergency preparedness and response; and an update on Chapter 16.1 and emergency response report-
ing.
Mine Environmental Control – the last session for the conference will discuss the following topics: Maximising
spot cooling potential when utilizing varying mine service water as a heat sink; an interactive training model;
and an innovative system for firefighting: Case studies highlighting the application to goaf heating mitigation
and a pro-active spontaneous combustion prevention technique.
The papers presented at the conference have been prepared by professionals in their field and have been peer-
reviewed according to the guideline used for papers published in the MVVSA Journal.
The MVSSA wishes to thank the keynote presenter and all authors including participants for their contributions.
As it is winter season, we urge all to adhere to personal hygiene, and be cautious of respiratory-related infec-
tions. We are excited to welcome you all to the 2023 MVSSA conference.
Constance Kekana
Conference Committee Chairperson
iii
The manuscript review committee consisted of the following members:

Ronald Mothlamme, Frank von Glehn, Dries Labuschagne, Marthinus van der Bank, Vijay Nundlall, Lance
Paiken, Marco Biffi, Selvin Subban, Schu Schutte, Tjaart Cronje; Stephan Bergh, Kobus Dekker, Wynand
Marx, Raymond Funnell, Jenny Branch, Kobus van Zyl
The organizing committee consisted of the following members:

Constance Kekana (conference chairperson), Johan Maass, Liezl Slabber, Madelein Terre ’Blanche, Marthinus
van der Bank, Cecilia Pretorius, and Johan Pienaar.
iv
Contents
Planning and designing an efficient mine ventilation network by FS Bergh 1
Tailoring plant engineering to fast-track supply and construction of refrigeration systems by A. Branch 9
Preparing a mines ventilation network to support expansion of the underground workings by FS Bergh
16
& S. Ambrosio
Investigating the business case for implementing EBM - a literature review on improving the overall
productivity of underground diesel fleets by JD Swanepoel, J. Hines, V. Gopaldasani, CJ Badenhorst 25
and Brian Davies
Women and ergonomic risk factors in underground coal mining. by OS Mokwena 33
Is carbon offset realistic for South African mining industry by Z. Jordaan 41
The role of artificial intelligence in enhancing mine ventilation, emergency preparedness, and occupa-
48
tional hygiene by F. Jacobs
Novel ionic liquid additive for silica scrubbing by R. Bosch and KG Schulz 60
A review of the application of Continuous Real-Time Monitoring in Noise Risk Management by K. 69

Ramoshaba
Establishing a process for the setting of real-time detection system alarms by RE Henderson 76
Real-time detection of Nitrogen Dioxide in mining atmospheres: issues, concerns, setting alarms, and
81
interpreting data by RE Henderson
Remote Safety Audits – Drawbacks and Hidden Treasures by K. Van Zyl, H. Botha, M. Biffi 88
Assessment of innovative mitigating solutions for fires occurring on conveyor belts in coal mines by 89
W. Marx, A. Thomson, M. van der Bank, N. van der Merwe
Emergency Evacuation Lessons Learnt by D. Van Der Walt, L. Van Wyk, M. Du Toit, D. Byleveldt 104
Maximising spot cooling potential when utilising varying mine service-water as heat sink by Dr. PvZ
115
Venter & Prof. M. van Eldik and J van der Merwe
Interactive Training Model by FT Maimane 123
v
Planning and designing an efficient mine ventilation network
F.S. Bergh
Howden Africa
Arnold Theron
Exxaro Resources
ABSTRACT: Mine Ventilation is fundamentally important to any underground mining operation. Despite this
fact, efficient mine ventilation design often does not enjoy the attention it deserves during the planning stages.
Historically, mine ventilation systems were designed with the end view in mind and mines ventilated at full
capacity for the duration of their life, and irrespective of the level of activity underground.
Recent energy price increases and commitments to net zero emissions are forcing mine planners to design more
optimised ventilation solutions. This paper sets out to sketch the economic landscape and describes ways to
improve the synergies and integration between mining plans, mine planning teams and efficient ventilation
systems to support a sustainable future for mining.
Amongst others, it engages topics such as the role of ventilation within the mine planning process, synergies
between departments, planning for changes in the legislative environment, the adoption of optimization tech-
nology and the strategic re-skilling of personnel.
Two case studies are briefly discussed to highlight the economic advantages of employing ventilation optimi-
zation solutions. The paper also recommends a framework for ventilation planning and optimization throughout
the mine planning cycle that enables the transition to more efficient and optimized ventilation strategies.
1. INTRODUCTION
Mine Ventilation forms an integral part of any mining operation. Despite the critical role it plays in worker
health and safety, as well as the sustainability of the operation, its importance is often underestimated, or worse,
it does not form part of the planning and design process. This results in inefficient and non- cost effective
ventilation strategies. (Clarke & De Souza, 2019)
Mining is a cornerstone of the South African economy. The mining sector in 2018 employed 453,543 people,
contributed R 22 billion in taxes in South Africa, as well as R 127.4 billion in employee earnings. The Minerals
Council (2018) estimates that the sector contributed 7.3% to the gross domestic product (GDP) in 2018, an
estimated contribution of R 356 billion.
The rising cost of electricity, along with a struggling Electricity Supply Commission (ESKOM) is casting a
dark shadow over the industry as mines become less profitable. The rapidly increasing input costs, threaten the
sustainability of the industry. “Electricity, steel and wage costs have risen much faster than inflation and elec-
tricity prices have trebled in the past seven years. Due to the combination of rand commodity prices stagnating
since their peak at the beginning of 2017, and costs rising consistently faster than the former, the mining sector
has been in a ‘profit squeeze’ for at least the last 5 years” (MINCOSA, 2018)
1
Developing a framework and integrating this with various stages of mine planning, as well as employing opti-
misation technologies can assist mines by supplying airflow of an acceptable quantity and quality to parts of
the mine where personnel and/or machinery is required to travel or work. In doing so, energy savings can be
realized whilst the health and safety of workers remains priority. (Clarke & De Souza, 2019)
Despite the obvious advantages of ventilation optimisation technologies, there has been an extremely slow
uptake within the South African Mining Industry. Pierce (2017) said it best when he mentioned that if new
technology continues to be sidelined, the effect of rising production costs, smaller profit margins and a reduc-
ing foreign direct investment will continue to be felt.
This paper hopes to provide interested stakeholders with an approach where ventilation and production planning
is integrated, and ventilation optimisation technologies applied to create sustainable ventilation strategies
The research methodology used was the desktop technique. The methodology used was qualitative in nature. It
typically includes literature reviews to explore different theories.
Section 1 provides background and introduces the idea of how optimized ventilation solutions can benefit
mines.
Section 2 focuses on the literature review. Topics such as ventilation within the mine planning process, syner-
gies between departments, planning for changes in the legislative environment, the adoption of technology and
the strategic re-skilling of personnel.
Section 3 focuses on the analysis of the reviewed literature and discusses some findings.
Section 4 presents conclusions.
Section 5 focuses on recommendations.
Section 6 lists all sources used for information gathering.
2. LITERATURE REVIEW
Mine planning boasts a variety of layers aimed at integration, risk identification and optimization. The purpose
of mine planning and the different stages involved in a mine plan, are to gauge the feasibility of a project. The
insights and knowledge gained, at varying and increased levels of accuracy throughout the different front end
loading stages, will assist stakeholders to gauge the feasibility of a project. It is during the planning stages that
information is assessed, risks and/ or challenges identified and strategies selected in order to optimize the eco-
nomics of a given project. (Shillabeer, 2001)
Mine ventilation considerations are often under-estimated, or even overlooked during these stages which lead
to inefficient and uneconomical designs. (Clarke & De Souza, 2019)
This literature review will explore the current planning process with respect to mine ventilation, and explore
ways to optimize mine ventilation strategies to ensure mines remain feasible and sustainable.
2.1 Background
There has been a weighted average increase in the cost of electricity from 1990 to 2016. The Chamber of Mines
indicates that, after Eskom’s 19,9% price adjustment, 66% of gold and platinum mines may be unsustainable
and 48 000 jobs could be lost” (The Chamber of Mines Nersa Submission, 2017)
2
The above statement calls for the planning and design of more efficient mines to ensure a sustainable future.
South African mines, even though designed from a worst-case scenario point of view, which is rather supply
too much ventilation or refrigeration as opposed to optimization enjoying preference, were always very lucra-
tive.
In the current economic climate however, the challenges posed are different. Designing a mine and applying
the brute force principle of ventilation was possible historically, however the points mentioned earlier make it
clear that safety must always remain top of mind, the design of a mine can no longer ignore all other facets.
(Bluhm, et al., 2001)
A safe, energy-efficient and optimized mine is the mine of tomorrow and the points listed above are forcing
mines to become more energy efficient in their approach. Failure to focus on this will render mines unviable,
and this will lead to higher rates of unemployment, losses in GDP and tax generation to only name a few.
Robust and integrated planning processes focused on optimized and efficient ventilation strategies, as well as
ventilation optimization technologies need to be considered.
Technological advancements, such as ventilation optimization solutions, are supported by Macfarlane (2001)
and Pierce (2017). Macfarlane concluded that, although high initial investment costs are the concern, mines can
actually reduce their operational costs when intelligent mines are developed.
Pierce commented that if new technology continues to be sidelined, the effect of rising production costs, smaller
profit margins and a reducing foreign direct investment will continue to be felt.
There are also those who feel that these technologies and their associated risks are too high and that the capital
investment and increased maintenance of autonomous devices are too costly and require reliable strategies,
which are not currently featuring in the South African Mining Industry.
As mentioned by Baiden et.al (2005) despite the benefits of ventilation optimization technologies, adoption has
been slow. This could be due to the demanding software requirements and significant initial capital costs, as
well as the re-skilling of personnel and maintenance requirements of the optimization systems.
These requirements are relatively minor in comparison to the benefits, and, in comparison to existing ventilation
systems, an optimized solution is clearly the better choice (Tanveer Jahir, 2011).
2.2 Ventilation Within the Mine Planning Process
The design of a mine ventilation plan often requires a number of assumptions to be made. These assumptions
are likely to change over the life of mine as the underground footprint expands, changes to the vehicle fleet and
possibly the mining method is introduced and portions of the underground workings become depleted.
Throughout these changes, it is imperative that the ventilation design complies with the relevant health and
safety legislation, whilst remaining practical, economic and efficient. (Clarke & De Souza, 2019)
Bluhm et al. (2001) noted that mine plans change and these revisions require ventilation considerations to ensure
the design remains efficient, whilst safeguarding employee health and safety. During these changes the venti-
lation engineer might be excluded, or in an attempt to ensure compliance, said engineer often designs excess
capacity to such an extent that the design can no longer be deemed efficient or economically optimized.
It is important that the level of detail and modelling, that informs the ventilation strategy, matches the different
levels of mine planning and that challenges or risks are identified during the feasibility, long-term and medium-
term plan before they are experienced underground. (Clarke & De Souza, 2019)
3
Clarke & De Souza (2019) recommended a framework to continuously test the ventilation strategy against not
only legal compliance and alignment to best practice, but also to ensure the system remains efficient whilst
doing so. Table 1 below is an adaption of this framework and the authors added their own comments based on
experience.
Table 1: Proposed Methodology for Efficient Ventilation Planning (Clarke & De Souza, 2019)
Phase Description
Inputs Simplify the Mining Plan: Number of available headings, number of active headings
(development and production), active vehicle fleet to achieve target tonnages, shift schedule.
Establish ventilation design criteria and determine the ventilation requirement to satisfy the
criteria.
Brownfields: Survey the underground workings taking (at least) pressure, quantity and tem-
perature measurements at representative places underground. Use these measurements to en-
sure the ventilation model is accurate and calibrated
Greenfields: Design the mine and construct the model to meet the benchmarked and legislated
design criteria in Step 3
Trouble Shooting Interrogate the calibrated model and identify airways/ parts of the mine that do not meet the
acceptable criteria.
Furthermore, the software to be utilised to identify bottlenecks , areas of excessive resistance

or pressure loss etc.
Optimise Once the constraints have been identified, test different scenarios or implementation plans to
address the constraints in the system.
( Study)
Conduct a thorough analysis of the options that ensure the acceptable criteria is met: consider
economics, health and safety as well as the environment
Optimise Engage Original Equipment Manufacturers (OEM’s) and investigate technologies and equip-
ment that cannot only supply the airflow during steady state, but also optimise the airflow
during “off-peak” periods.
Examples include: Fans with Variable Speed Drives (VSDs) so that flow rates, hence energy
expenditure, can be reduced during non-production windows such as travelling times or
maintenance periods.
Another example is to investigate Ventilation on Demand technologies, where a tracking so-

lution or instrumentation underground delivers the desired demand only where, and when,
vehicles or personnel are active
2.3 Mine Ventilation Planning Throughout the Life of Mine

One of the main reasons for inefficient ventilation systems, is that design, of specifically the primary ventilation
network, is often carried out and based on the airflow requirements when the mine is fully developed and pro-
ducing at maximum capacity. This long term view to ventilation planning is important, but ignoring or not
4
considering part load during the ramp up phase of the mine leads to highly inefficient systems. (Papar, et al.,
1999)
Whilst the long-term plan is strategically focused and vitally important, ignoring the different planning horizons
and devising a ventilation strategy that only supports the long term view, will have a negative impact on the
economics of the project. Short term operational planning, otherwise referred to as tactical planning, and must
support the operation to be optimized throughout in order to favorably impact the metrics of success, be it from
a health and safety or economic point of view. Devising a long term ventilation strategy, that aligns with the
mining plan, and testing the said strategy at shorter intervals to see its influence on the future strategy is best
practice and should form part of an operations planning cycle on an annual basis. (Kloppers, et al., 2015)
2.4 Synergies and Interaction Between Departments

It is of utmost importance that departments communicate, discuss and compromise during the planning process.
This team approach should continue to exist during the life of mine and seek compromise or alternatives in
order to satisfy conflicting demands. Practical examples could include shaft sizing or the enlarged development
of a critical airway underground, which might not make financial sense nor completely satisfy the rock engineer,
but will ensure ventilation is provided to a part of the workings for the duration of the mine.
The regular and ongoing liaison between departments and ensuring a multidisciplinary approach will assist
mines to avoid expensive redesigns, whilst ensuring an optimized design and layout. (McPherson, 1993)
2.5 Planning for Changes in the Legislative Environment
Clarke and De Souza (2019) noted that Mine Ventilation Engineers are tasked with designing environments
that ensure worker health and safety, whilst being energy efficient and optimized. As changes in legislation,
with regards to acceptable limits of exposure to contaminants, become more stringent ventilation strategies also
need to adapt to ensure statutory requirements are adhered to. It is therefore of utmost importance that Ventila-
tion Professionals form an integral part throughout the planning cycle of a mine. Regular updates of the mining
plan and ventilation strategy deem it necessary that a multi-disciplinary task teamwork towards achieving this
common goal. A prime example of this would be the recent movement and implementation of standards around
diesel engines and Diesel Particulate Matter (DPM) exposure in Underground Mines (Stinnette, 2013)
2.6 The use of Technology and Strategic Upskilling of Personnel
Mine planning involves people, process and technology. The process needs to be defined and managed, people
upskilled and the best available technology put to use in order to achieve the desired outcome. (Smith, 2022) In
recent years, the use of Ventilation Simulation Software has become a powerful tool used by Ventilation Pro-
fessionals to plan, design and optimize ventilation networks. Increased production rates, greater mining depths
and more extensive excavations means more complex ventilation systems. (Bluhm, et al., 2001)
The increased complexity, and competitive nature of mining, will require optimized ventilation solutions. This
will necessitate the evaluation of numerous options and scenarios. Scenario planning will also be required once
mine layouts change to ensure the design is appropriate, compatible, safe, efficient and cost-effective. This
process would be almost impossible without the aid of simulation tools that are able to rapidly evaluate, in a
flexible and accurate way, the different scenarios and options. (Bluhm, et al., 2001)
Reflecting on the statement above, the strategic re-skilling of employees is critically important. Ventilation
professionals need to be educated and empower themselves to utilize the available technologies. The synergies
between people, process and technology offers Africa the opportunity to increase our competitiveness globally,
whilst the introduction and adoption of technologies should be seen as an opportunity, not a threat. The previous
5
industrial revolutions aided in job creation, and through proper human resource planning and development, the
fourth will be no different. (Viljoen, 2018)
3. RESULTS AND ANALYSIS
3.1 Overview
This section aims to showcase how two mining operations have benefited from adopting this integrated ap-
proach. This illustrates an approach where a mine ventilation network is optimised through multi-disciplinary
planning, as well as the implementation of technology to create a safe and efficient system.
3.2 Findings and Analysis
Newmont’s Eleonore Gold Mine employs a ventilation optimization system that allows them to manually, or
autonomously, control their underground ventilation devices. The devices can react based on schedules, feed-
back received from environmental sensors or based on the location of vehicles and personnel within the under-
ground workings. The focus on creating a safe, but energy efficient ventilation system has led to a 43% reduc-
tion in mine heating costs. Furthermore, the secondary ventilation system has noted a 56% reduction in
electricity cost, whilst a decrease of 73% in electricity cost was recorded on the Primary Ventilation Equipment.
(Gleeson, 2019)
Another example was the optimization of a mine’s cooling system as described by Bluhm et al. (2001) where
scenario planning, through ventilation simulations, identified an opportunity to reduce the operations power
consumption by 11%, or R 185 million.
4. CONCLUSION
Planning and designing a ventilation strategy for a mine is critically important. Mine ventilation directly impacts
the health and safety of the workforce, as well as the economics of a project. To add onto this, mining houses
are making Environmental, Social and Governance (ESG) commitments to ensure a sustainable future. Venti-
lation optimization forms a key part of this strategy, as an optimized ventilation solution can assist mines to
significantly reduce their energy consumption, and subsequently carbon footprint. (Bluhm, et al., 2001). Alt-
hough ESG is high on the radar of mining houses currently, the concept of not only measuring a business based
on its financial metrics, but also the impact it has on people and the environment was identified by before the
turn of the millennium. (Elkington, 1998)
Mine ventilation designs are often done with a long-term view in mind. As discussed by Clarke & De Souza
(2019). Kloppers et al. (2015) noted the importance of synergies between the long term and short term or tactical
plan. Failure to test the strategy at different stages of the planning process, or incorporate ventilation planning
at different planning horizons , could lead to inefficient designs in the shorter term. A common oversight, as
noted by Papar et al. (1999) mines often don’t pay enough attention to the energy efficiency of their ventilation
systems.
In conjunction with the integration of ventilation planning at different stages of the mine planning process,
mines need to employ technologies and solutions to aid them in devising efficient ventilation strategies. As
mines become deeper and more complex so do does their ventilation systems. Bluhm et al. (2001) discussed
how available technologies can support ventilation engineers in designing efficient systems. The coming to-
gether of people, process and technology is key to a mine’s ventilation optimization strategy. The examples
given of how mines can benefit from employing more intelligent ventilation technologies are showcased in this
6
document. Strategically re- skilling personnel to utilize technology and work in an environment to deliver a
solution where people are supported by machine learning and advanced analytics is the cornerstone of digital
transformation journey. (van den Berg, 2022)
5. RECOMMENDATIONS
Based on the information contained in this document it is evident that ventilation planning needs to play a key
role in the mine planning process. The design of solutions that are efficient throughout the life of the project
must be considered.
Recommendations include developing a system where ventilation planning and mine planning interrogate the
efficiency, and applicability, of the ventilation strategy on frequent basis. This is to ensure the long-term venti-
lation plan supports the current mine, as well as actions to address the future needs of the project.
Re-skill ventilation engineers to utilise the available technologies and perform feasibility studies focused on
mine ventilation specifically. These studies must identify and explore ventilation optimization technologies,
their advantages and associated costs. As with other facets of mine planning, these studies to focus on total cost
of ownership and establish the feasibility of the various solutions against metrics such as net present value,
internal rate of return and payback period as opposed to performing a simplistic payback period calculation.
Government must support mining houses that explore more sustainable and environmentally friendly ventila-
tion solutions. This can be enabled in a number of ways, for example by incentivizing mines that invest in these
technologies by offering more compelling electricity tariffs. Another means of supporting the transition would
be to necessitate mines to publicly report their ventilation optimisation initiatives.
Mines need to develop a framework that reconciles the ventilation and the product plan. This can be achieved
by testing constraints in the system against acceptable criteria based on law, regulations and best practice. A
gap analysis must then be performed and ventilation modelling employed to ultimately deliver a solution that
meets the needs of not only an economic analysis, but also considers issues such as people and the environment.
(Clarke & De Souza, 2019)
6. REFERENCES
1. Bluhm, S., Marx, W., Von Glehn, F. & Biffi, M., 2001. VUMA MIne Ventilation Software. Cracow, Bluhm
Burton Engineering; CSIR Miningtek.
2. Clarke, R. & De Souza, E., 2019. www.airfinders.ca. [Online] Available at: https://www.airfinders.ca/wp-con-
tent/uploads/2021/07/A -Pragmatic-Methodology-for-Mine-Ventilation-Planning [Accessed 9 August 2022].
3. Elkington, J., 1998. Accounting for the Triple Bottom Line. Measuring Business Excellence, 2(3), pp. 18-22.
4. Eskom, D. S. M. D., 2010. The energy efficiency series: Towards and energy efficient mining sector, Johan-
nesburg: Eskom Holdings Limited.
5. G Baiden, Y. B. S. H. K., 2005. Researchgate. [Online] Available at: https://www.researchgate.net/publica-
tion/300395195_Justification_of_on-demand_ventilation_systems_by_means_of_simulation_techniques [Ac-
cessed 28 September 2020].
6. Gleeson, D., 2019. International Mining. [Online] Available at: https://im-mining.com/2019/11/07/howdens-
eleonore-ventilation-on-demand-solution-wins-award/ [Accessed 12 August 2022].
7. Kloppers, B., Horn, C. & Visser, J., 2015. Strategic and tactical requirements of a mining long term plan. The
Journal of the South African Institute of Mining and Metallurgy, 115(6), pp. 515-521.
7
8. McPherson, M. J., 1993. Subsurface Ventilation and Environmental Engineering. 1st ed. Dordrecht: Springer
Netherlands.
9. Minerals Council South Africa, 2018. Facts and Figures Pocket Book. 1st ed. Johannesburg: Minerals Council
South Africa.
10. Papar, R. et al., 1999. Missouri University of Science and Technology. [Online] Available at: https://scholar-
smine.mst.edu/usmvs/8usmvs/8usmvs-theme16/3/ [Accessed 12 August 2022].
11. Shillabeer, J., 2001. Lessons Learned Preparing Mining Feasibility Studies. In: A. Edwards, ed. The Australian
Institute of Mining and Metallurgy Guide to Good Practice. Melbourne: The Australian Institute of Mining and
Metallurgy, p. 435.
12. Smith, G., 2022. Mine Planning Principles : Risk in Mine Planning. Johannesburg: University of the Witwaters-
rand.
13. Stinnette, D. J., 2013. Establishing Total Airflow Requirements for Underground Metal/ Non-metal Mines
Based on the Dieself Equipment Fleet, Kingston, Ontario, Canada : Queens University .
14. Van den Berg, J., 2022. Planning Considerations [Personal Communication] (20 August 2022).
15. Viljoen, L., 2018. What is the fourth industrial revolution?. [Online] Available at: https://www.iol.co.za/busi-
ness-report/opinion/what-is-the-fourth-industrial-revolution-14127465 [Accessed 03 August 2019].
8
Tailoring plant engineering to fast-track supply and construction of re-

frigeration systems
A. R. Branch
BBE Projects
ABSTRACT: Northam Platinum Zondereinde No.1 Shaft has an extensive refrigeration system, producing
chilled water for use in surface bulk air coolers and as service water underground. A plant extension was
urgently required to supplement the water-cooling capacity. The brief included construction and
commissioning as quickly as possible, without affecting the operation of the existing plant. This paper
describes how changing the engineering approach to the plant design allowed the extension work to be fast-
tracked. Some key areas were: selection of more readily available materials and equipment; incorporating
design features to protect sensitive equipment and increase longevity of lighter materials; maximizing off-
site activities to run concurrently with site work, in order to deliver equipment closer to ready-for-
commissioning and reducing site activities. The result was that the plant extension was constructed and
commissioned in eight months, notably quicker than accepted norms.
1. INTRODUCTION
1.1 Existing surface refrigeration plant

The existing surface refrigeration plant at this mine was constructed in stages between the late 1990s and 2004,
and consists of eight York centrifugal water chillers, operating on R-134a. The plant produces chilled water
which is used in a bulk air cooler on surface and by various consumers underground, including hydropower
equipment and underground cooling equipment. An underground refrigeration plant was added in 2013 to ser-
vice a bulk air cooler. A key operating philosophy is to return the service water from underground to surface as
warm as possible, thereby extracting the maximum possible cooling from the service water circuit. The heat
rejected from the condensers of the underground chiller is also added to the return service water. Once on
surface, the return service water is passed through pre-cooling towers, which provide low-cost atmospheric
cooling ahead of the existing water chillers.
9
Table 1. Original thermal design parameters for the pre-cooling tower

______________________________________________________
______________________________________________________
Parameter Value
______________________________________________________
Total cooling capacity 56 MWR
Number of pre-cooling tower cells 8 off
Total service water flow through tower 1 300 l/s
Pre-cooling tower water inlet/outlet temperatures 30,0 / 19,6 °C
______________________________________________________
Table 2. Original thermal design parameters for the existing refrigeration plant
____________________________________________________
Parameter Value
______________________________________________________
Number of refrigeration machines 8 off
Total service water flow through evaporator circuit 1 300 l/s
Pre-cooling tower water inlet/outlet temperatures 15,0* / 5,0 °C
______________________________________________________
Note: Water off the pre-cooling tower is blended with water returning from the surface bulk air cooler, and a portion of chilled water
is back-passed around the chillers to achieve the design inlet temperature to the machines.
The desired chilled water temperature off the existing plant for distribution underground and to the surface
bulk air cooler is 5 °C.
Over time, the performance of the pre-cooling towers has been affected by various factors:
• operating efficiency was diminished through fouling caused by water quality circumstances.
• Increased heat load from the underground refrigeration machines added to the water returning to surface,
leading to elevated water temperature entering the pre-cooling towers, above the original design of 30 °C.
• the cooling tower design, from 30 years ago, was based on a lower ambient air wet bulb temperature than
commonly used today: 18 °C vs 20 °C.
The condenser cooling towers that service the refrigeration machines have also been affected by similar fac-
tors.
These factors caused the temperatures of the water entering both the condensers and the evaporators of the
refrigeration machines to be elevated, which had two effects:
• the chillers ran at higher-than-design condensing temperatures and pressures, with commensurate increase
in compressor absorbed power, costing the mine more to operate, and often limiting the machines’ cooling
output, and
• the desired outlet chilled water temperature of 5 °C could not be achieved as the entering temperature was
too high and the chillers are not rated for the elevated delta T. The elevated chilled water temperature
meant that the mine was not receiving full benefit of the cooling produced in the refrigeration plant. The
elevated water temperatures could also cause the chillers to overload and trip, reducing cooling to the
mine and placing more strain on the operating units.
10
Three projects were launched to improve the refrigeration plant performance, to ensure operational stability
and to increase the overall cooling capacity:
• replace the pre-cooling tower fill to restore the original operating efficiency and cater for the elevated inlet
water temperature from underground,
• replace the condenser cooling tower fill to improve the existing refrigeration machines’ efficiencies, and
• install additional refrigeration capacity.
The different options available for the additional refrigeration capacity were:
• extend the existing plant with same refrigeration machines to provide installed stand-by capacity,
• build more pre-cooling towers to maximise atmospheric cooling prior to service water entering the existing
chillers, or
• build an in-line additional refrigeration system between the pre-cooling towers and the existing refrigeration
plant to mechanically chill water before it enters the existing chillers
The decision was taken to add an in-line cooling system with new water chillers, as this would provide the best
operational flexibility and allow for maintenance outages and break-downs in the existing plant. Simply adding
more machines to the existing plant would not solve the problem of bridging the thermal gap (in summer)
between the pre-cooling circuit and the existing machines, it would simply result in the back-passing of even
more water. Furthermore, the water would have to be produced at the same 5 °C, which would be far less
efficient than intermediate machines producing water at 15 °C. Similarly, increasing the number of pre-cooling
towers would also not tackle the thermal gap issue during periods of elevated ambient wet bulb temperatures.
1.2 New Cooling System Thermal Performance

The thermal design parameters for the new additional cooling system are:
Table 2. Thermal design parameters for the additional cooling system

______________________________________________________
Parameter Value
______________________________________________________
Number of refrigeration machines 4 off
Total service water flow through evaporator circuit 1 000 l/s
Pre-cooling tower water inlet/outlet temperatures 18,1* / 14,5 °C
______________________________________________________
Note: blended temperature off pre-cooling towers and return BAC water.
The thermal design parameters were specified at the revised peak summer conditions to ensure the existing
chillers would receive water into their evaporator heat exchangers at a maximum temperature of 14,5 °C.
The service water heat load does not vary widely between summer and winter as the temperature of the hot
water returning from underground remains relatively constant. However, the pre-cooling tower performance
does increase in colder months, lightening the load on the mechanical refrigeration equipment. In the shoulder
seasons and winter, the new chillers will be set to maintain their maximum cooling duty and relieve the existing
chillers, as the new chillers will be more efficient.
11
1.3 New Cooling System Design Concept

The new cooling capacity project was conceived with the following primary features:
• quick to construct and commission,
• integrate with existing plant with the minimum disruption to existing equipment,
• as compact as possible – due to limited real estate,
• be able to operate entirely independently from the existing equipment,
does not have to have a very long design life as No. 3 Shaft refrigeration plant would be commissioned by
end 2024 and would assist with overall mine cooling demands.
Figure 1: Process Flow Diagram for Additional Cooling System
The key areas to consider in order to fulfil this concept included:

• selection of most readily available materials and equipment
• incorporate design features to protect sensitive equipment and increase longevity
• maximise off-site activities to run concurrently with site work, in order to deliver equipment closer to
ready-for-commissioning and reduce site activities
2. NEW COOLING SYSTEM DESIGN FEATURES
2.1 In-line construction

The plant was constructed in an open area outside the main mine boundary, completely separate from the
existing pant, thereby having no effect on existing plant’s operation.
The only tie-in points necessary were into pipes on the outlets of the pre-cooling towers to divert water to
the new plant, and then return it downstream in the same pipe. The tie-ins were completed over an off-weekend
when the mine could afford to shut the existing plant down.
12
2.2 Service water isolation from refrigeration machines

In order to isolate the new refrigeration machines from the potentially corrosive mine water, four plate heat
exchangers arranged in parallel were used, with evaporator circuit running in a closed-loop between the cold
side of the heat exchangers and the chillers.
Titanium was selected as the plate material for its superior corrosion resistance properties. The closed-loop
configuration allowed the use of copper tubes in the shell-and-tube evaporators. Copper tubes have a better heat
transfer coefficient, are cost effective and, importantly, readily available which contributed to a shorter produc-
tion time compared with using a special material such as copper-nickel alloy.
One disadvantage of using plate heat exchangers is the risk of fouling due to suspended solids in the service
water getting lodged in the narrow spaces between the plates. To counter this, strainers were installed upstream
of the heat exchangers to catch larger debris items. Multiple strainers and heat exchangers were installed in
parallel which means they can be individually isolated and cleaned while keeping the new plant operating at
full flow by increasing the water flow through the remaining units.
Figure 2: Plate heat exchanger bank with upstream strainer
2.3 Condenser cooling tower selection

The cooling towers selected for this project are film-pack packaged units. These have two distinct advantages:
• they are fully factory assembled, then split for transport. Installation on site consists of joining the two
halves of each tower and is quick, and
• film pack has the best heat transfer characteristics of all types of cooling tower fill, which means the cooling
towers are compact relative to the amount of heat they can reject. The draw-back to this fill is that it does
require better quality make-up water and is prone to fouling when used for heat rejection duty, consequently
needing replacement more regularly than other types of fill.
13
Figure 3: Factory-assembled package cooling towers
2.4 Substation construction

The electrical substation is constructed from two converted shipping containers, arranged side-by-side,
which are fitted out with medium and low voltage switchgear, PLC panel and ancillary equipment (small power,
lighting etc) at the vendor’s works. The entire substation was inspected and tested in the factory to ensure as
far as possible that all equipment functioned correctly and all communications between the various components
were correct. This was all completed while the electrical contractor focused on field equipment construction
(racking, cabling etc). When the substation was delivered, all that remained before site commissioning was for
cables to be terminated to the MV and LV panels, and between the Control & Instrumentation panels.
This approach was especially useful as the refrigeration machine OEM provided soft-starters to feed the
chillers, which had to be integrated with the MV switchgear and PLC control signals.
While the use of converted containers is gaining popularity for its convenience, one particular challenge is
to ensure that electrical regulations are complied with, for example, minimum space requirements for access
and exit are adhered to.
3. CONCLUSION
This project has achieved its stated goals of accelerated construction and commissioning. However, while this
lightweight, accelerated concept was a valid solution for this project, every cooling system has its own unique
requirements and challenges which necessitates its own tailor-made design.
14
Figure 4: Aerial view of site during construction
15
Preparing a mines ventilation network to support expansion of the

underground workings
F.S. Bergh
Howden Africa
S.K. Ambrosio
Howden Ventsim (previously Vortex OHS)
ABSTRACT: This specific operation envisioned ramping up their production and plans included a deepening
project to extend the life of the operation. This entailed a review of their current ventilation network, as well as
steps to be taken to ensure the workings remain conducive throughout the period of expansion.
The operation was looking for a solutions partner that could support their drive towards zero harm. The project
comprised of the following work:
• Ventilation survey of the underground workings.
• Using this data, construct a calibrated Ventsim™ model of the workings.
• Use the calibrated model to design a robust ventilation system that will ensure underground conditions
comply with global best practice and support the company’s commitment to zero harm.
• Complete an optimisation assessment and identify short-, and medium-term actions to support a safe
and healthy working environment.
The proposal ensured the mine remained adequately ventilated to create workplaces which are safe and risk
free to the health of employees. A few of the highlights included:
• A 19% increase in Total Mine Air Utilisation.
• A 32m³/s increase in airflow on the lowest level of the workings, following phase 1 of the staged im-
plementation plan. This will ultimately increase to 60m³/s following phase 2.
• A 6.2 ºC reduction in wet bulb temperatures on the lowest level of the mine. This was realised without
the addition of any mine cooling equipment.
• An 83% reduction in airborne pollutant contaminant levels on the lowest level of the workings
1. INTRODUCTION
This specific mine has been in operation since 1966 and the workings extend approximately 1.2 km below
surface. Following a plunge in nickel prices in 2008, the mine began operating on "care and maintenance.”
Today the mine is operational again and plans include a deepening project to extend the life of the operation.
As a result of their depth, geothermal properties of the host rock, age of the operation and age of ventilation
infra-structure the underground conditions posed challenges to their future plans of deepening the current work-
ings.
16
The operation was looking for a solutions partner that could support their drive towards zero harm, by designing
and implementing a mine ventilation system that would enable and support their endeavours and plans for
expansion. Howden was contracted to complete the following work:
• Ventilation survey of the underground workings.
• Using this data, construct a calibrated Ventsim™ model of the workings.
• Use the calibrated model to design a robust ventilation system that will ensure underground conditions com-
ply with global best practices and support the company’s commitment to zero harm.
• Complete an optimisation assessment and identify short-, and medium-term actions to support a safe and
healthy working environment.
The design ensured the mine remained adequately ventilated to create workplaces that are safe and rrisk-freeto
the health of employees. A few of the highlights included:
• A 32m³/s increase in airflow on the lowest level of the workings, following phase 1 of the staged implemen-
tation plan. This will ultimately increase to 60m³/s following phase 2.
• A 6.2 ºC reduction in wet bulb temperatures on the lowest level of the mine. This was realised without the
addition of any mine cooling equipment.
• An 83% reduction in airborne pollutant contamination levels on the lowest level of the workings
2. VENTILATION MODELLING: THE CALIBRATED MODEL

To ensure that the Ventilation Simulations/ Ventilation Models accurately reflect the mining operation, great
care was taken to ensure the base parameters were accurate and based on proven research. This included, but
was not limited to the following:
• Base energy cost
• Surface environmental conditions
• Ground conditions and host rock properties
• Airway dimensions and friction factors
• Resistances of ventilation appliances
• Heat sources
The operation furnished Howden with ventilation plans of the operation. This information, as well as on-site
measurements taken by Howden personnel, were used in conjunction with the Computer Aided Design (CAD)
files of the underground workings.
It was often necessary to make assumptions as to the condition of ventilation appliances as these were not
always stipulated in the references received.
The information provided by the operation was of a sufficient nature that we were ultimately able to achieve
balance of the model with airflows within 5% (and often closer) of those values indicated on the ventilation
plan. The calibrated model can be seen in Figure 1 below:
17
Figure 1: The Calibrated Model
2.1 Analysis of the calibrated model

In summary, the mine is supplied with 225.8 m³/s at its upcast shaft at an annual energy cost of $1 461 407.
This results in a ventilation unit cost of $6 472 per m³/s.
A total of 76.4 m³/s (33.8% of all available ventilation) is provided to Production Commitments at their main
returns. The sections themselves are ventilated by means of auxiliary fans.
A further 50.0 m³/s (22.1% of all available ventilation) is provided to Non-Production Commitments (i.e., the
workshop and dust raise) at their main returns.
Therefore, in total 126.4 m³/s is utilised by these commitments. On this basis, a total system air Utilisation of
56% is achieved, with wasteful leakage accounting for 99.4 m³/s of the total upcast capacity.
3. SIGNIFICANT FACTORS DETRIMENTAL TO THE CURRENT SYSTEM
3.1 Primary Fans

The state of the surface fans was quite concerning, and the most pertinent issue identified was damaged dampers
within each fan drift. Numerous damper vanes were not in the fully open position, and this resulted in an un-
necessary pressure loss of ±2500 Pa.
The high pressures, and increased system resistance caused by the damaged dampers indicated that the fans
were operating at approximately 46% efficiency.
3.2 System Leakage

Leakage within any ventilation system is always problematic. During the course of the underground surveys
conducted, numerous incidences of leakage into the return and abandoned areas was observed.
In the case of the latter, given the age of the workings and the unavailability of comprehensive survey plans, it
was not always clear where this air was flowing to. Irrespective, there was an obvious argument to seal off all
old workings and to address any leakages directly to the return. This also being advantageous from a safety
point of view, as it prevents persons entering abandoned areas
18
3.3 Pressure Loss within the Return Airway System

A pressure survey was conducted along the main return airway of the mine.
A total of 2925 Pa is made available to the system by the surface fans. This is best described as the available
force to overcome the resistance of the workings and allows ventilation to flow within them.
On Level 33/1, the available pressure drops to a negligible -25 Pa. There are two key observations that should
be made from this:
• This means that the surface fans are effectively being used to overcome the resistance of the area
above Level 33/1 to surface, and;
• The remainder of the workings (i.e., lower than 33/1) is ventilated by some other mechanical means,
as air movement is observed on these lower levels.
It is, at this point, necessary to appreciate the structure of the main Return Airway Raise.
There are 3 significant portions to the raise:

1. From Surface to Level 9, the return consists of a 4.0m x 4.0m (16.0 m²) steeply inclined airway (30°
from the horizontal).
2. From Level 9 to Level 35, the return consists of a series of Ø5.0m (19.635 m²) near-vertical raises.
The raises appear to have been blasted and not raise-bored. This means they have a relatively high
friction factor.
It is important to note that these raises are offset from one another. This results in small transition airways,
connecting each raise which cause substantial shock losses due to the change in direction.
3. From Level 35 downwards, use is made of a variety of Ø2.0m (3.142 m²) vertical raises.
In certain locations as many as 4 raises are used between levels and at others, as little as 1 raise has been used.
From level 37 downwards, only two raises with a combined area of 6.284 m² are used. The Return Airway
Raise is therefore severely constrained by this bottleneck. The return airway system is depicted in Figure 2
below
Figure 2: Return Airway System
19
The changes in size, which varies from a maximum on 19.635 m² to as little as 3.142 m², provides additional
severe shock losses to the system.
Using the main return line from surface to the lowest portion of the mine, this return has a combined length of
≈1970m of varying dimensions.
The main Return Airway Raise has a combined resistance of 0.1431 Ns²/m⁸. Considering that the surface fans
experience a resistance of 0.1923 Ns²/m⁸ without the resistance of the dampers (as explained in the Surface Fan
Section), this means that the entire balance of the system has a resistance of 0.0492 Ns²/m
Based on this, 75% of all fan energy expended on the system is being used to overcome the main Return Airway
Raise.
Worse, if cognisance is taken of the resistance currently posed by the dampers, 86.5% of all fan energy is
expended to overcome the main Return Airway Raise and the Fan Dampers, leaving only 13.5% to ventilate
the rest of the mine.
Booster Fan Installations

Later, in the survey, it was determined that the other issue detrimental to the system is the “mini” booster fan
installations that were installed at various production levels to boost the return airway. It should be noted that
these booster installations perform 2 functions.
Firstly, they move the Zero Static Pressure point further up the main return of the mine. In this case it is just
below Level 33/1.
Secondly, they create backpressure, which tries to push out of the return airway system back into the intake
system. Later surveys by the mine confirmed this and the lower system is effectively recirculating. An example
of one these recirculation loops is shown in Figure 3 below:
Change size of fig 3 and include caption below
Figure 3: Recirculation Loop
As mentioned above, the system is recirculating. In addition, there is insufficient ventilation of the lower work-
ings to provide for sufficient dilution of heat and airborne pollutants. Figure 4 below illustrates the severity of
the problem, with airways in pink and purple exceeding 32.5 Temperature Wet- Bulb
20
Figure 4: Wet Bulb Temperatures
4. SHORT TERM IMPROVEMENT PLAN

The Short-Term Improvement Plan (STIP) aims to improve the efficiency of the existing system. Numerous
significant changes to the system are required.
It is important to appreciate that the section that follows is not intended to be comprehensive (as it is difficult
to show all changes).
4.1 Repair of primary fans

Briefly, this required the operation to attend to the previously mentioned challenges at the primary fans.
Such measures included the following:
• Repair the dampers in each fan drift. Theoretically, this should result in an increase of 2500 Pa to the
rest of the system.
• This alone should move the pressure point discussed the lower than Level 33/1.
• Conduct a full engineering assessment of the surface fans and make any required remedial measures to
allow fans to perform as per original design.
4.2 Sealing off Abandoned Areas in the Old Workings

As discussed previously, a significant amount of ventilation (which is difficult to quantify) moves through the
older workings. This poses significant risk, not least of which is leakage possibly directly into the main return.
It was strongly recommended that the operation do the following:

• Systematically seal off each abandoned area on each level.
• Due regard should be given to ensure that access to the main Return Airway Raise is accessible for the
purposes of measurement and maintenance.
21
An example of the layouts supplied for each level can be seen in figure 5 below:
Figure 5: Ventilation Layout depicting Changes.
5. ANALYSIS OF THE SHORT-TERM IMPROVEMENT PLAN

In summary, the mine will now be supplied with 237.8 m³/s (up by 22.0 m³/s) at its up cast shaft at an annual
energy cost of $1 513 726. This results in a ventilation unit cost of $6 366 per m³/s.
A total of 107.0 m³/s (45.0% of all available ventilation) is provided to Production commitments. A further 65.9
m³/s (27.7% of all available ventilation) is provided to Non-Production Commitments.
Therefore, in total 172.9 m³/s is utilised by Commitments. On this basis, the following system air utilisation is
achieved:
• Total Air Utilisation = 72.7%
• Wasteful leakage accounts for 64.9 m³/s of the total upcast capacity.
Most importantly, however, is the fact that all “mini” booster installations have been removed and that 32.0
m³/s now flows naturally to the lowest level.
This resulted in a significant reduction in heat and pollutant build-up in the workings.
6. MEDIUM TERM IMPROVEMENT PLAN

The Medium-Term Improvement Plan (MTIP) aimed, once again, to improve the efficiency of the existing
system by making use of a Booster Fan installation in the middle region of the mine.
These changes are significant from an improvement and cost perspective. It is important to appreciate that the
section that follows is not intended to be comprehensive.
6.1 Booster fan installation

Firstly, a Ø5.0m Raise Borehole needs to be established from Level 35/5 to Level 35. This will form the basis
of a new return to the lower levels.
22
This new raise bore will hole into the existing Ø2.0m raise borehole. This smaller raise borehole was previously
mined as part of an intended expansion of the main Return Airway Raise.
Preliminary indications are that this raise will need to be reamed to Ø5.0m in any future expansion of the main
Return Airway Raise.
Secondly, an additional airway connecting to the main Return Air Raise needs to be mined. The size of this
airway is at the operations discretion but should be able to accommodate the size of the proposed booster fans
Modelling suggested that 3 x Ø1600 315kW Dual Stage - Full Blade Booster Fans & Columns be installed.
Two Fans will be operational, while the third will be used as a spare (and possibly even operational in the Life
of Mine).
These two fans combined should providean up cast capacity at this location of ± 175.0 m³/s.
Preliminary indications are that with all 3 fans operational, and without any changes to the network, these fans
will deliver a combined up cast capacity of ± 205.0 m³/s.
Figure 7 below shows the layout changes required in this area.
Figure 6: Level 35 Ventilation Layout Changes
With the installation of the Booster Fans complete, the amount of ventilation on Level 45/2 (lowest level) is
anticipated to increase from 32.0 m³/s (in the STIP) to 60.0 m³/s.
The balance is being used to ventilate other production levels. This is a substantial improvement, which allows
for more heat and pollutant dilution at the lowest level.
7. ANALYSIS OF THE MEDIUM-TERM IMPROVEMENT PLAN

In summary, the mine is now supplied with 269.0 m³/s (up by a further 31.2 m³/s) at its upcast shaft (with the
assistance of the Booster fans at Level 35) at an annual energy cost of $2 628 307. This results in a ventilation
unit cost of $9 771 per m³/s.
23
A total of 139.9 m³/s (52.0% of all available ventilation) is provided to Production Commitments, whilst
A further 61.8 m³/s (23.0% of all available ventilation) is provided to Non-Production Commitments.
On this basis, a system air utilisation of 75% is achieved
8. CONCLUSION
This paper has described the Short and Medium-Term Improvement Plans, as well as ventilation layouts and
arrangements that need to be put in place to ensure adequate ventilation is provided to all production sections.
The design ensured the mine remained adequately ventilated to create workplaces, which are safe and risk free
to the health of employees. A few of the highlights included:
• A 32m³/s increase in airflow on the lowest level of the workings, following phase 1 of the staged implemen-
tation plan. This will ultimately increase to 60m³/s following phase 2.
• A 6.2 ºC reduction in wet bulb temperatures on the lowest level of the mine. This was realised without the
addition of any mine cooling equipment.
• An 83% reduction in airborne pollutant contamination levels on the lowest level of the workings
24
Investigating the business case for implementing EBM - a literature

review on improving the overall productivity of underground diesel
fleets
J Deon Swanepoela,b*, Jennifer Hinesa,b, Vinod Gopaldasania,b, Cas J. Badenhorsta,b,c,d and Brian Daviesa,b
a
University of Wollongong, Faculty of Arts, Social Sciences and Health, School of Health and Society, Wol-
longong, NSW 2522, Australia
b
Centre for Occupational Public Environmental Research in Safety and Health (COPERSH)
c
North West University, Occupational Hygiene and Health Research Initiative (OHHRI), Potchefstroom, 2520,
South Africa
d
Anglo American Corporate Services South Africa (Pty) Ltd, Marshalltown, Johannesburg, 2107, South Africa
*
Corresponding author: E-mail address: jds986@uowmail.edu.au J Deon Swanepoel
ABSTRACT:
Diesel fuelled vehicles liberate diesel engine exhaust (DEE) that poses a carcinogenic health risk for exposed
individuals. Emissions-based maintenance (EBM) is a systematic diagnostic process to control DEE, by incor-
porating DEE and engine data to influence maintenance decision making. It is a source control tool with the
potential to benefit occupational health and hygiene risk reduction while simultaneously benefiting operational
production needs. Available research shows evidence for productivity gains such as reduced DEE at the source,
reduced fuel consumption, reduced worker exposure, and anecdotal evidence for extended exhaust aftertreat-
ment (EAT) service life. There was no evidence that EBM improved fleet management (measured as fleet
availability and reliability) or resulted in reduced underground dilution ventilation delivery. A research project
is currently underway at the University of Wollongong to delineate EBM intervention and routine planned
maintenance practices and to investigate the effect of EBM on fleet management. No EBM related research
from the South African Mining Industry (SAMI) was found despite the large underground diesel fleets in oper-
ation. It is recommended that SAMI consider EBM and routine planned maintenance practices as critical pre-
ventative control measures.
INTRODUCTION
Emissions-based maintenance of diesel fleets has the potential to benefit both production requirements and air
quality of work environments, especially in the underground mining industry. Diesel engines are commonly
used for heavy-load transportation due to their design and strength, unfortunately emit carcinogenic diesel en-
gine exhaust (DEE) posing a health risk for exposed workers (IARC 2012). Diesel engine exhaust is a complex
mixture of hundreds of hydrocarbons, polycyclic aromatic hydrocarbons, submicron diesel particulate matter
(DPM), and inorganic gases, including carbon monoxide (CO), nitrogen oxides (NOx), and sulphur dioxide
(U.S. EPA 2002; IARC 2012).
The power output of a diesel engine is largely determined by the amount of fuel that is combusted within the
engine cylinders. Lean operation is commonly employed in diesel engines to facilitate complete combustion of
fuel, whereby an excess of air is supplied. Conversely, a fuel-rich condition, which is characterized by insuffi-
cient air for complete combustion, can result in elevated levels of incomplete combustion products such as CO,
DPM, and other combustion by-products (Holtz 1960). Over time, emissions quality can decline due to gradual
wear and tear on engine components, leading to a rise in concentrations of diesel emissions. Furthermore, some
engine malfunctions can act synergistically to exacerbate emissions, such as intake or exhaust restrictions in
tandem with over-fuelling (Waytulonis 1992a). Nevertheless, appropriate maintenance practices can mitigate
the degradation of emissions (Waytulonis 1992a).
Emissions-based maintenance is a planned maintenance process that involves the collection of engine emissions
data to assess an engine's combustion and exhaust aftertreatment (EAT) efficiency relative to baseline data or
similar engine types. This process is used to make targeted decisions to reduce emissions or prevent impending
engine failure or faults. A statistical method for identifying high-emitting engines involves comparing an indi-
vidual engine's emission data to the 95% upper confidence level of an identical fleet of engines (Davies, 2004).
Engines with emissions exceeding this level should be investigated for deviating from the fleet's emission
25
profile. This process includes quantifying raw and treated diesel exhaust emissions (DEE), measuring pressure
differentials across control technologies, and inspecting six engine systems: 1) intake air system; 2) engine
cooling system; 3) diesel fuel handling and quality; 4) fuel injection system; 5) lubrication system; 6) exhaust
system (McGinn 2000; Holtz 1960; Waytulonis 1992a). Emissions-based maintenance consists of four key
procedures: (1) periodic monitoring of engine systems; (2) data-based engine fault diagnosis; (3) targeted
maintenance to address diagnosed faults; and (4) confirmation testing to evaluate intervention effectiveness
(Bugarski et al., 2004).
Although the Mine Health and Safety Act and Regulations promote worker health and prevent occupational-
related harm, this legislation does not regulate diesel emission or DPM exposure. Therefore, it is up to opera-
tions to determine to what extent they can reasonably control the health risks introduced by diesel fleets. Inter-
national legislation for diesel engine maintenance and -emissions exist in countries such as Australia, Canada
and the United States (Rangal 2019).
This literature review focuses on the advantages associated with the implementation and adherence to an EBM
program. Productivity gains are defined as any quantifiably positive outcome from a production and occupa-
tional health perspective. These gains are summarized under the following categories: reduced EAT-out emis-
sions, reduction in fuel consumption, improved diesel fleet management, extended EAT service life, reduced
worker exposure, and reduced dilution ventilation delivery underground. Engine-out emissions refer to exhaust
constituents measured at the exhaust manifold, and exhaust aftertreatment-out (EAT-out) emissions refer to
measurements after emission control technology.
METHODOLOGY
A literature search was conducted to investigate productivity gains related to EBM. The search was con-
ducted across multiple databases including Web of Science, Scopus, Computers and Applied Sciences Com-
plete, Health Business Elite, MasterFILE Complete, CINAHL Plus, Information Science and Technology, and
Environment Complete. Four unique search syntaxes, each covering a different theme relevant to the investi-
gation of productivity gains related to EBM, were created using key words and Boolean operators. The themes
included diesel engines, maintenance, diesel engine exhaust, and occupational health and safety and included
full text peer-reviewed, and grey literature publications up to 1 March 2022.
In addition to the electronic search, a manual search was conducted of all available Mining Diesel Emissions
Council (MDEC) conference presentations and the Centres for Disease Control (CDC) and Diesel.net webpages
for "emissions-based maintenance" and "emission-assisted maintenance program".
The process of selecting publications for this review is presented in PRISMA Diagram. (Figure 1). Initially,
4996 documents were identified through the search process. Following the removal of duplicates, 2113 docu-
ments underwent applicability screening. From this, 84 documents were assessed in full text for eligibility. No
academic journal publications that investigated the effects of maintenance informed by DEE data were found
for inclusion in this review. However, eight publications in the grey literature that pertained to DEE and
maintenance were deemed relevant and were included. The benefits presented in these eight publications are
summarised in Table 1.
26
Figure 7: PRISMA Diagram
27
RESULTS
Table 1 summarizes the productivity gains and the quantitative evidence reported in the eight publications in-
cluded in this review. Five of the publications are based on in-situ case studies by industry experts and all of
the publications are from grey literature. The publications do not provide information to compare EBM fault
diagnosis with site specific planned maintenance.
All eight publications reported reduced emissions resulting from EBM informed maintenance decisions, which
included interventions on the air intake, fuel, and exhaust system of engines (Davies, 2004; Davies and McGinn,
2005; Forbush, 2001; Forbush, 2006; McGinn, 2000; McGinn et al., 2002; Hines, 2019; Volkert and Rhiley,
2007).
Three publications reported deviances in the air intake system requiring maintenance (McGinn, 2000; McGinn
et al., 2002; Davies and McGinn, 2005). These interventions, including replacing air filters and addressing air
leaks, led to various degrees of reduction in measured DEE. Four publications reported reducing maximum fuel
delivery to reduce emissions (Davies and McGinn, 2005; Forbush, 2006; Forbush, 2001; Volkert and Rhiley,
2007), while one publication increased the fuel delivery after another maintenance was completed to regain
some of the lost power output (Davies, 2004). Seven publications reported reduced emission after maintenance
was conducted on EAT devices, which included using emission data to identify poor performing EAT devices
and to rationalise fitting different EAT devices (Davies, 2004; Davies and McGinn, 2005; Forbush, 2001;
McGinn, 2000; McGinn et al., 2002; Hines, 2019; Volkert and Rhiley, 2007).
Davies (2004) reported that the replacement of fuel injectors in two separate engines, informed by out-of-spec-
ification engine control data, resulted in a significant reduction in exhaust emissions. The first engine showed
a 55% reduction in elemental carbon (EC) and an 18% reduction in CO, while the second engine had a 40%
reduction in EC and a 30% reduction in CO. In another scenario, a high back pressure resulting from a blocked
scrubber tank led to high EC and CO concentrations. After chemically treating the scrubber tank and fitting a
new fuel pump, there was a 67% reduction in EC and a 45% decrease in CO. Fuel injectors were replaced in an
attempt to improve engine power, which had virtually no effect on the EC concentration but did further reduce
CO by 55%. Furthermore, reducing the fuel delivery rate based on the high CO concentration resulted in a 66%
reduction in raw exhaust EC and more than a 72% reduction in CO.
McGinn (2000) provides case studies with five different engines where visual inspections, engine data and DEE
data were used to identify and correct engine faults. The author emphasizes the importance of maintaining the
intake system, which was found to be deficient in three of the engines. By addressing the intake system, CO
and EC were reduced in the range of 34-49% and 6-59%, respectively. By addressing all identified faults, the
average reduction at the tailpipe was 49% for CO. Exhaust after treatment efficiency, specifically diesel
oxidative catalysts for CO mitigation, improved from 40% to 69% after faults in the exhaust system were
corrected.
In order to assess the transferability of the process outlined in the Diesel Emission Evaluation Program Report
(McGinn 2000), McGinn and colleagues (McGinn, Huzij et al. 2002) conducted three case studies in which an
EBM approach was utilized to inform targeted maintenance practices. The authors report significant reductions
in emissions of NO, NOx, and DPM (reported as smoke), ranging from 33% to 40%, as well as a notable
increase in power following maintenance actions such as cleaning the catalytic converter, replacing injectors,
adjusting valves, and addressing intake restrictions.
28
Table 1: Summary of productivity gains associated with EBM.
Gains stated Publication Quantitative Results
Reduced EC (%) Reduced CO (%) Reduced NOx (%)

i. Davies (2004) i. 40 – 67 i. 18 – 72 i. Not reported
ii. Davies and McGinn
ii. 41 - 67 ii. Not reported ii. Not reported
(2005)
iii. Forbush (2006) iii. Not reported iii. 90 iii. 50
Reduced EAT- iv. Forbush (2001) iv. Not reported iv. 90 iv. 50
out emissions v. Hines (2019) v. 91 v. 25 v. 66 (as NO2)
vi. McGinn et al., (2002) vi. 33 - 50 vi. 50 - 65 vi. 35 - 75
vii. McGinn (2000) vii. 17 - 66 vii. 36 - 60 vii. 8 - 59
viii. Volkert and Rhiley
viii. 95 viii. 98 viii. 75
(2007)
Reduced personal i. Statistically significant decrease of 33% in EC.
i. Hines (2019)
exposure Oxides of nitrogen inconclusive
i. Forbush (2006) i. 23% reduction in fuel usage
Reduced fuel
consumption ii. 15% reduction in fuel usage (20% and 6% respective for
ii. Hines (2019)
LHDs and PTVs)
i. Reduced maintenance cost accredited to increased engine
i. Forbush (2006
life from 3,000 hours to 10,000 hours
ii. Forbush (2001) ii. Not quantified*
Improved fleet
management iii. Not quantified*
iii. Volkert and Rhiley *Publications reported productivity gains associated with fleet
(2007) management, framed as reduced downtime, reduced break down,
or increased engine life
i. Increased average time of 90 hours before filters needed
i. Davies (2004)
Improved EAT change
service life ii. highlighted that premature disposable diesel exhaust filter
ii. Forbush (2006)
change out occur
i. Forbush (2006) i. Not quantified**
ii. Forbush (2001) ii. Not quantified**
Reduced dilution
ventilation iii. Not quantified**
iii. Volkert and Rhiley
**Publications did not provide quantified reduction in dilution
(2007)
ventilation.
DISCUSSION
The most widely cited benefit of EBM is the reduction in DEE, which can be divided into a decrease in
engine-out emissions and a reduction in EAT-out emission (Davies, 2004; Davies and McGinn, 2005; Forbush,
2001; Forbush 2006; McGinn, 2000; McGinn et al., 2002; Hines, 2019; Volkert and Rhiley, 2007).
Several studies indicate that EBM data can be used to inform maintenance decisions for the intake-, fuel-,
and exhaust systems, or a combination of these systems, to result in reduced EAT-out emission (McGinn, 2000;
Hines, 2019).
Back pressure and reduction efficiency are important factors in EAT. Simple tasks such as unblocking ex-
haust scrubber tanks can reduce excessive back pressure, increasing control efficiency and reducing CO and
DPM. Addressing exhaust system damage and leakages can also improve EAT reduction efficiency. Therefore,
reducing engine-out CO and DPM can contribute to the reduction of these EAT-out pollutants.
It should be noted that the reporting of the effects of targeted maintenance on different constituents of DEE
may not be entirely causal. For instance, improving the air-fuel ratio by maintaining the intake system is ex-
pected to reduce both CO and DPM compared to pre-intervention data. However, this effect may be
29
counteracted to varying degrees when other maintenance activities are conducted, such as alterations to the fuel
system that increase fuel delivery.
A case study by McGinn (2000) showed that maintaining the intake system resulted in the anticipated re-
duction of CO and DPM at the manifold, but subsequent alterations to the fuel system and engine adjustments
increased DPM content by 13%. These nuanced increases highlight the complexity of the interaction between
DEE constituents and maintenance, and that EBM does not inherently lead to a correlated reduction of all
pollutants for all interventions. The nett effect was still a reduction in emissions. Interactions between produc-
tivity gains exist, such as reducing back pressure to decrease diesel usage, and optimizing fuel delivery for lean
combustion to decrease CO and fuel consumption but are complex and beyond the scope of this review. Suc-
cessive testing can optimize engine power output while meeting emission targets, as demonstrated by Davies
(2004). Similarly, the reduction in power due to reduced fuel delivery can be compensated for by changing
torque converters (Forbush 2006; Forbush 2001).
Studies by Hines (2019) and McGinn (2000) have shown that the reduction of engine-out CO and DPM
substantiates claims of improved fuel consumption. Additionally, lean combustion has been shown to reduce
CO and DPM in diesel engine exhaust by Holtz (1960), resulting in improved fuel utilization and increased
power output per volume of fuel. Furthermore, the reduction of engine-out EC by Hines (2019) and McGinn
(2000) can partly explain the extension of filter life. A higher DPM content in the exhaust stream will lead to
quicker filter occlusion and subsequent increased filter changes, resulting in earlier increases in backpressure.
More frequent filter changes will directly increase maintenance cost.
The evidence suggests that emissions-based maintenance can benefit production needs in terms of fuel cost
savings as well as occupational hygiene and health requirements. However, there is currently no evidence sup-
porting productivity gains for fleet management.
The reviewed publications do not provide evidence that EBM has led to a reduction in dilution ventilation
in underground mines. However, it can be reasonably assumed that reducing emissions at the source, particu-
larly from high emitting engines, could result in a reduction in dilution ventilation requirements. There is a need
to test this hypothesis in future research.
Investigating the impact of an EBM program on fleet availability and reliability could expedite industries to
formally implement such programs for operational and occupational health gains.
It remains unclear if EBM will yield the same value for mechanically and electronic diesel engines, and there
is a need for a diagnostic tool to assist relevant personnel with engine fault diagnostics.
CONCLUSION
This paper investigated the benefits of adopting EBM in underground coal mines. It reviewed various studies
that have reported EBM associated productivity gains, including reduced diesel emissions, reduced personal
exposure to DPM, improved fuel economy, improved fleet management, increased EAT service life, and re-
duced ventilation requirements. The available literature arises from grey literature i.e. not from peer reviewed
literature, and are often derived from case studies by industry experts. Consequently, any extrapolation of
productivity gains associated with EBM is limited. Emissions-based maintenance data can be used to inform
maintenance decisions and reduce engine-out emissions for CO, DPM, and NOx. Furthermore, there is evidence
that indicates that controlling emissions at the source can reduce personal exposure to DPM. Finally, there is
evidence that the adoption of EBM can lead to fuel consumption reduction.
Diesel emission constituents are inter-related and influenced by various factors, which can make engine fault
diagnostics a complex and challenging process. Currently, there is no widely available central database to assist
with interpreting exhaust emissions data.
Consequently, engine fault diagnostics rely on operator skill, knowledge and experience. A research pro-
gram at the University of Wollongong is currently underway to construct a diagnostic tool based on EBM data
to facilitate engine fault diagnostics.
30
No EBM related research from the South African Mining Industry were found despite the large underground
diesel fleets in operation. It is recommended that SAMI consider EBM and routine planned maintenance prac-
tices as critical preventative control measures.
REFERENCES
Bugarski, A., G. Schnakenberg, J. Noll, S. Mischler, M. Crum, and R. Anderson. 2004. "Evaluation of diesel
particulate filter systems and biodiesel blends in an underground mine." In, 207-18.
Bugarski, Aleksandar D., Jr George H. Schnakenberg, Jon A. Hummer, Emanuele Cauda, Samuel J. Janisko,
and Larry D. Patts. 2009. 'Effects of Diesel Exhaust Aftertreatment Devices on Concentrations and Size
Distribution of Aerosols in Underground Mine Air', Environmental Science & Technology, 43: 6737-
43.
Campbell, John D, Andrew KS Jardine, and Joel McGlynn. 2016. Asset management excellence: optimizing
equipment life-cycle decisions (CRC Press).
Canada Occupational Health and Safety Regulations (COHSR). 2014. "Part X – Hazardous Substances." In,
edited by Canada Occupational Health and Safety Regulations.
Davies, B. 2004. 'The Control of Diesel Particulates in Underground Coal Mines', Victoria University of
Technology.
Davies, B., and S. McGinn. 2005. "The Effect of Maintenance of Diesel Engines on Particulate Generation." In
IOHA.
Forbush, S 2001. "Diesel Emissions Reduction Program." In Mining Diesel Emissions Conference. Markham,
Ontario, Canada.
Forbush, S. 2006. "Emissions-based maintenance program." In, 111-13. Taylor and Francis/ Balkema.
Health and Safety Executive (HSE). 2012. "Control of Diesel Engine Exhaust Emissions in the Workplace." In
HSG 187, edited by HSE.
Hines, J. 2019. 'The role of Emissions Based Maintenance to reduce diesel exhaust emissions, worker exposure
and fuel consumption', University of Wollongong.
Holtz, John C. 1960. Safety with mobile Diesel-powered equipment underground (US Department of the
Interior, Bureau of Mines).
International Agency for Research on Cancer (IARC). 2012. "IARC monographs on diesel and gasoline engine
exhausts and some nitroarenes." In, edited by International Agency for Research on Cancer. Lyon,
France: WHO.
Kothamasu, Ranganath, Samuel H. Huang, and William H. VerDuin. 2006. 'System health monitoring and
prognostics - a review of current paradigms and practices', Int J Adv Manuf Techno, 28.
McGinn, S. 2000. "The relationship between diesel engine maintenance and exhaust emissions." In. Sudbury,
Ontario: DEEP.
McGinn, Sean, Bob Huzij, and Dave Cisyk. 2002. "Diesel Emissions Maintenance Training Program." In
Mining Diesel Emissions Council.
Mine Safety and Health Administration (MSHA). 1914. "Mandatory Safety Standards. Underground coal
mines. Maintenance of diesel-powered equipment." In 30 CFR 75, edited by Office of the Federal
Register. Washington, DC: U.S. : Government Printing Office. 1988. "Diesel engines intended for use
in underground coal mines, 30CFR, Part 7, Subpart E." In, edited by MSHA.
New South Wales (NSW). 2014. "Work Health and Safety (Mines) Regulation." In.
New South Wales (NSW) Department of Primary Industries. 2008. "MDG 29 Guideline for the management
of diesel engine pollutants in underground environments Mine Safety Operations Division." In, edited
by Maitland NSW Department of Primary Industries, Australia.
Raposo, H., J. T. Farinha, I. Fonseca, and L. A. Ferreira. 2019. 'Condition monitoring with prediction based on
diesel engine oil analysis: A case study for urban buses', Actuators, 8.
United States Environmental Protection Agency (U.S. EPA). 2002. "Health Assessment Document for Diesel
Engine Exhaust (Final 2002)." In, edited by Office of Research and Development U.S. Environmental
Protection Agency, National Center for Environmental Assessment. Washington Office, Washington,
DC, EPA/600/8-90/057F.
31
Volkert, S, and T Rhiley. 2007. "Practices and principles associated with DPM compliance at Queenstake
Resources." In DPM Workshop. Reno, Nevada.
Waytulonis, R. 1992a. "The Effects Of Maintenance And Time - In - Service On Diesel Engine Exhaust
Emissions." In, edited by United States Department of the Interior. U.S. Bureau Of Mines. 1992b. "An
Overview Of The Effects Of Diesel Engine Maintenance On Emissions And Performance." In, edited
by United States Department of the Interior. U.S. Bureau Of Mines.
32
Women and ergonomic risk factors in underground coal mining.
Authors: OS Mokwena and Dr TJ Morodi Tshwane University of Technology Pretoria South

Africa, Dr J Shirinde University of Pretoria South Africa
ABSTRACT: Mining operations are known for their challenging working conditions, and underground mining
in particular is regarded as one of the most physically taxing jobs. The task's physical requirements, the sur-
rounding environment, and information obtained via displays are additional ergonomic risk factors. A compre-
hensive strategy for minimizing musculoskeletal injuries among female mineworkers must include a mine-
specific ergonomics program built on participative principles. The purpose of the article is to identify the current
situation and disclose any gaps in the ergonomics risk factors impacting women working in South African
underground coal mines. The current knowledge on ergonomics in underground mining and the associated risk
factors that influence female mineworkers was examined using a systematic literature review. Lower back dis-
comfort and musculoskeletal disorders are the most often reported injuries and the main reason for absenteeism.
The aforementioned absenteeism is frequently brought on by routine work tasks like carrying heavy equip-
ment’s or sitting in a chair that isn't right for one's physique.
1. INTRODUCTION
Globally, women have become an essential part of the mining workforce, mainly because of equal
opportunity legislation, which obliged mining companies to actively change the demographic pro-
file of their workforce (Botha, 2016). The number of women working in the mining sector, coal
mines included, has increased significantly in the past 15 years or so, from around 11,400 in 2002
to around 53,000 women in 2015, increasing to 54,154 in 2018 (Baxter, 2019). Globally, women
constitute more than 10% of the large-scale mining workforce (Rickard et al., 2017), while in
Ghana, female participation in large-scale mining is estimated at 10% (Bansah, Dumakor-Dupey
& Yalley, 2017 & Khan, 2013). In the Unites State (US) for instance, women constitute 13% of
mining workforce (Mines Magazine, 2019) which is predominantly large-scale, this demonstrate
that more women are entering the mining sector and as a result, the sector ought to ensure that the
environment and the systems in place, considers the safety of underground working women as
well. Women represent 12% of South Africa’s mining labour force, from a total of 454,861 female
mineworkers, 13,059 (14%), are in coal mining (Mineral Council South Africa, 2020). The newly
elected South African democratic government of 1994 adopted a number of strategies to open up
the mining sector to historically disadvantaged South Africans, which include women, as part of
its economic empowerment policy. Among others, new mining legislation, such as the Mineral
and Petroleum Resources Development Act (No. 28 of 2002) and the Broad-based Socio-economic
Empowerment Charter for the South African Mining Industry (the Mining Charter) (RSA, 2010),
was introduced. These pieces of legislation not only prohibit the exclusion of women but also
require from companies to actively change the demographic profile of their workforce by setting
specific targets to be reached by 2015 (Harmony Gold Mining Company, 2008). The South African
administration, is currently reviewing policies to ensure equal participations of everyone, particu-
larly women in the mining industry (Kandare, 2017). This regulatory effort is in line with United
Nation's Sustainable Development Goal (SDG) 8; to achieve gender equality and empower all
women and girls. Mining has always been considered a very masculine industry due to its heavily
male-dominated workforce as well as the physicality of mining work (Botha & Cronje, 2015), the
nature of working in a mine, and in particular underground, is deemed hazardous and the related
33
jobs require a high degree of physical strength and endurance (Bansah, Dumakor-Dupey & Yalley,
2017).
The actual workplaces in deep level mines present challenges to ergonomics interventions, mining
takes place in very restrictive work areas with low ceiling heights and high thermal heat loads and
still involves a large component of physical work despite the introduction of engineering measures
and mining equipment intending to make work easier (Steven,2013). Like other areas of
occupational health and safety (OHS) ergonomics is evolving and becoming more integrated into
overall work management systems. The Ergonomic Regulations (2019) defines ergonomics as “the
scientific discipline concerned with the fundamental understanding of interactions among humans
and other elements of a system, and the profession that applies theory, principles, data and methods
to design in order to optimise human well-being and overall system performance”. Mining
equipment used in the narrow reefs is designed for use without considering basic ergonomics
principles and female mineworkers. The typical ergonomics related hazards associated with
trackless mining equipment underground include restricted vision from the driving position,
restricted cabin space, difficult vehicle access for the operator and exposure to whole body
vibration, noise and dust (Duarte, Marques & Baptisa, 2021). Technological changes in the mining
sector and mining methods fuelled by the need to excel and produce more has seen the traditional
manual mining methods being replaced by mechanisation which exposes miners to sedentary
routines in equipment operation. This in particular, increases the exposure of operators of load
haul dump (LHD) trucks and risk of injury due to postural environments and whole-body
vibrations (Dube & Chiluba, 2021). Given that ergonomics efforts in mining aim to prevent
musculoskeletal disorders by reducing exposure to awkward postures, high force exertions, and
task repetition. Unfortunately, underground work environments exacerbate exposure and make
ergonomics efforts more challenging for female mineworkers than in other industry sectors
(NIOSH, 2019). Therefore, it is important to go beyond the concept of ergonomics studies centred
on single, static workplaces, which may be valid for offices and computer stations but not for most
miners, and women who move through different workstation in underground mining (Apud &
Meyer, 2009).
2. METHODOLOGY
The information that is currently available on ergonomics in underground mining and the associated
risk factors that affect female mineworkers was examined using a systematic literature review
methodology. In addition to data sources including PubMed, Medline, Science Direct, Web of
Science, and Google Scholar, grey literature, such as reports was used. Studies published in English
between 2000 and 2022 that at least "examine women in underground mining" and "ergonomic risk
factors impacting female mineworkers" met the requirements for eligibility. The results from many
studies were combined using the meta-analysis approach. For a paper to be shortlisted, it was essential
to have at least the keyword(s) or one of the keyword combinations (mining ergonomics, female mine
workers) in its title or abstract. The manuscripts were carefully studied to include only those fulfilling
the criteria that the papers must contain studies reporting the prevalence of ergonomic risk factors
among female mineworker globally. The literature review aimed to present an overview of ergonomic
risk factors impacting female mineworkers in underground coal mines, as well as the gaps in research
related to female mineworkers in underground coal mining, South Africa.
34
3. RESULTS
Findings according to Coal-tech 2020, indicate that males and females have biologically different
physiques. This has a significant influence on how females execute certain tasks as required per
job category (Ashworth et al.,2004). In the Coal-tech 2020 study, research compared males with
females and it was determined that females have a lower manual handling capacity. The implica-
tion of this finding is that the majority of female mineworkers are incapable of lifting and pulling
items of a certain weight (Ashworth et al.,2004). In another study conducted by Injury Prevention
Control Australia (IPCA), they collected data on more than 900 men and women, in one of the
largest comparative studies of the sexes. Each of the study participants performed 10 leg exten-
sions with the heaviest weight load possible. This strength assessment for the front thigh muscles
(quadriceps) was conducted on a specific leg extension machine equipped with a special computer
to ensure proper exercise technique with respect to movement speed as well as full movement
range. The results revealed that the male participants were about 50 percent stronger than the fe-
male participants. According to the researchers if they divide the weight, what men and women
lifted, by their body weight, to adjust for weight differences, the average male completed 10 leg
extensions with 62 percent of his body weight and the average female completed 10 leg extensions
with 55 percent of her body weight (Heine, 2012), this further proves that the manual material-
handling capabilities of women are substantially lower than those of men, attributed primarily to
differences in muscle strength (Schutte, 2010).
The European Foundation's European Union-wide survey found that women were more likely than
men to undertake repetitive tasks or repetitive work (Banerjee et al., 2016), and Bello (2020) em-
phasise a similar trend and revealed that women are suffering because of the arduous, repetitive
jobs they do in underground coal mines. In the case of carpal tunnel syndrome, which caused more
days away from work per case than any other workplace injury or illness, more women sustained
carpal tunnel syndrome by operating machinery, than they did typing, keying, and performing
other duties associated with office work (Livine et al., 2001). In Poland the effects were more
dramatic still. The statistics, broken down by gender, show reports of strain injuries, were 43 per
cent more common in women, at 3.3 cases per 100,000 employees compared to 2.3 for men. The
study found that out of 49 per cent of workers who were exposed to musculoskeletal constraints,
45.8 per cent were males and 63.2 per cent were females. Of these, 42.4 per cent of women, almost
two-thirds (60.3 per cent) were exposed to repetitive movements; 5.8 per cent to forced postures
(Cheng and Kawachi, 2000).
According to Shariat et al., (2016), musculoskeletal disorders as well as lower back pains are the
most reported injuries and the largely subsequent cause of absenteeism. The said absenteeism is
often caused by everyday work activities such as sitting in a chair not suited for one’s body or
lifting heavy objects using an improper lifting technique. Simultaneously, Ma et al. (2018), also
found that these injuries account for one in every three worker-reported injuries. Thus and so,
musculoskeletal disorders cause damage to the worker’s muscles, nerves, blood vessels, ligaments,
or tendons, and this can further include carpal tunnel syndrome, rotator cuff injuries, trigger finger,
tendinitis, epicondylitis, and strained muscles as hypothecated by Carmondy (2017). However,
what is disconcerting is that women have had to work with equipment or machines initially de-
signed for men and no provision was made at the time to cater for the needs of the women within
the mining sector. Pursuant to that, some of the underground machines are enormous, and in many
instances, may be designed with higher climbing steps, making it burdensome for women to mount
or descend off the machines or equipment (Matshingane, 2017). Additionally, gender differences
35
also exist regarding the aerobic capacity of men and women. Ashworth et al., (2004), defines aer-
obic capacity as the maximal oxygen uptake that provides a quantitative measure of a person's
ability to sustain high-intensity physical work for longer than five minutes. This implies that aer-
obic capacity of women is typically 15 to 30% below the values of their male counterparts. This
means that women work closer to their aerobic capacity than men and are thus more likely to
become fatigued. Fatigue is operationally defined as the 'reduced muscular ability to continue an
existing effort' (Ashworth et al., 2004,). High levels of fatigue can reduce performance and
productivity in the workplace and increase the risk of accidents and injuries occurring (Schutte,
2010). Research of MSDs amongst miners in South Africa has focused on the male-dominated
worker populations with Dias (2014) studying miners working in coal, gold and platinum mines
in 2014, reporting an annual MSD incidence of 2.7% (Dias, 2014). Although great efforts have
been made to accommodate women in the South African mining industry since the inception of
the Mineral and Petroleum Resources Development Act (MPRDA) in 2002, the literature reviewed
indicates the following main barriers that female mineworkers continue to face. Frequently, the
physical strength required for many mining tasks precludes female mineworkers from performing
these tasks effectively (Minerals Council South Africa, 2020). Difficult jobs such as operating rock
drills, locomotives, and dozers may hold physiological risks for female mineworkers (Mangaroo-
Pillay & Botha, 2020) According to Kabongo and Naidoo (2021), South African female min-
eworkers, reported that their biggest challenge was the heaviness and weight of machines and the
mine equipment which makes it difficult for them to operate the equipment. Working conditions
can further complicate this issue. In situations where female’s toilet facilities are situated far away
from places of work, female mineworkers may refrain from drinking while on shift, resulting in
dehydration – a dangerous safety risk (MCSA, 2020). Additionaly, Zungu (2011) propounded that
there is a need to increase awareness and knowledge of the 'safe limits' for female mineworkers
handling mining equipment and tools. Furthermore, female mineworkers cannot use the same tech-
niques as their male counterparts to lift and handle heavy objects and materials because of their
body size as well as body build which should be taken into consideration when appointing females
in core underground mining positions.
4. DISCUSSIONS
Short bursts of high energy-consuming work, punctuated by rest periods or lesser energy-
consuming duties, are the hallmark of contemporary coal mining. It has already been established
that there are differences in the aerobic capacities of men and women, henceforth, female
employees may perform differently from male employees since they have lesser aerobic capacities
and tend to become fatigued faster in the shift (Botha & Cronje, 2015). Although women and men
are both affected by machine design and ergonomics in the mining industry, women are more
likely to have issues with surviving these conditions. From a broader overview, the manual
material-handling capabilities of women are substantially lower than those of men, attributed
primarily to differences in muscle strength (Schutte, 2010). Given this difference, the ergonomic
features of such mining equipment appear not to provide for the physiological make-up of South
African female mineworkers. At the time when mines started to recruit women, equipment’s were
still only suitable for men, resulting in both women and men having to share such equipment.
Botha (2017) states that dimension is an important concept that should be taken into cognisant in
the design of mining equipment as well as its efficacious operation. Currently, this is a problem in
the mining sector because only a specific number of people are involved in the procurement of the
36
machinery, not including the people who will use them like female mineworkers. James and Scott
(2009) support
the aforementioned by omitting the fact that one of the main challenges has been the absence of
organization and coordination of efforts to implement ergonomics, which has led to the
implementation of no strategy for programs in the mines. Notwithstanding the good intentions of
the introduction of women into the very male, ultramusculine mining environment created, it still
creates challenges, not only for managers but also for male mineworkers and the newly employed
female mineworkers (Botha & Cronjé, 2015). Undoubtedly, the issues of machinery design for
operations and maintenance and whole-body vibration exposures when operating machines and
vehicles are becoming more critical. The link between prolonged sitting, poor cab design and
vibration with back and neck pain is being recognized but has yet to be addressed in any systematic
way by the mining industry (Kalkis et al., 2015).
Accordingly, Zungu (2011) propounded that there is a need to increase awareness and knowledge
of the 'safe limits' for women for handling mining equipment and tools. Furthermore, women
cannot use the same techniques as their male counterparts to lift and handle heavy objects and
materials because of their body size as well as body build which should be taken into consideration
when appointing women in core underground mining positions. It is necessary to keep moving
forward from diagnostic research to participative interventions, as a result, all new investments in
plants, no matter how big or little, must consider pertinent ergonomic principles from the
beginning of the design process. Therefore, it is critical to expand ergonomic training within
businesses for everyone involved, not just the decision-making managers but also the employees
who suffer directly from a lack of ergonomic management as well as mining medical practitioners.
The Ergonomic Regulation (2019) stipulates that medical monitoring be carried out and records
be kept; this merely means that mining medical practitioners must also be qualified to identify
ergonomic injuries at an early stage, which is already a challenge in the mining industry. Pursuant
to the Mine Health and Safety Act, and the Occupational Health and Safety Act, 1993 (Act No. 85
of 1993) the employer is required to provide and maintain a work environment that is safe and
sound without the risk to the health of employees. Risk assessments need to be done and the risk
of exposure needs to be assessed through risk identification of all specific jobs, to determine what
is potentially hazardous to the health of workers (OHSA, 1993). Furthermore, section 21 (1)(c) of
the Mine Health and Safety Act requires designers and manufacturers of mining equipment and
machines to ensure that their products comply with ergonomic principles (MHSA, 1996).
Eftimie et al. (2009) remind us that it is not enough for the extractives sector to simply employ
greater numbers of women, there are more structural issues that need to be addressed throughout
the industry in terms of gender equality, particularly in relation to appropriate safety equipment
and ergonomic. It is further recommended that all interested and affected parties (manufacturers
and suppliers) be involved in the practical implementation of the proposed ergonomics
interventions and conceptual designs. This is essential in terms of Section 21 (1) (c) of the Mine
Health and Safety Act (Act 29 of 1996) (MHSC, 2017).
5. CONCLUSION
Overall, we see a picture emerging of women mineworkers, however, women's involvement as
discussed above, also carries serious health consequences. Such health consequences stem from
the hazardous working conditions experienced by women mineworkers, the inadequate attention
given to health and safety, and the highly physically demanding nature of the types of tasks that
women commonly undertake in relation to mining ( Lahiri-Dutt, 2008). Accordingly, reducing
37
occupational injuries and ill health is a strategic objective of all the stakeholders in the South
African mining industry. With the view of achieving this objective, it is ineluctable to dispense the
safest and healthiest work environment possible for all mine workers, men as well as women
(Pelders & Ridders, 2020). It is commonly accepted that the application of sound ergonomics
principles in the design of mining tasks and the human-machine interface will minimize design-
induced human error, and will ipso facto also eliminate significant occupational health and safety
risks. In summary, the application of ergonomics tends to support the maintenance of health,
effectiveness, and well-being, particularly for female mineworkers. It is also imperative and
crucial that changes ought to be made in systems currently in operation to address ergonomic
deficiencies as they are detected. There thus an urgent need to increase the awareness of workplace
issues and their impact on Women in Mining (WIM) with the objective of promoting women's
Occupational Health and Safety (OH&S) in the mines, in particular, musculoskeletal strain (Apud
& Meyer,2012). A mine specific ergonomics program based on participatory principles, is an
essential component of a comprehensive plan for preventing musculoskeletal injuries amongst
female mineworkers. While the mining industry appears to be committed to alleviating many of
these problems, there is still much work to be done to implement and evaluate ergonomics risks.
The application of ergonomics principles can play an important role in reducing workplace hazards
and improve the design of work which will benefit workers of all ages and experience levels (de
Arrudaa & Gontijob, 2012). Designers can benefit from attending relevant occupational health
platforms. More needs to be done to protect mineworkers, particularly female mineworkers, as the
underground mining environment is still far from optimal. Though the situation is far from ideal,
The Minerals Council of South Africa has developed a white paper focused on streamlining
industry strategies to advance women in mining. It aims to do this by encouraging female
representation in the industry and to drive decisions that are in the best interest of women, which
one of the focus areas will be to recognising and adjusting for the different physical capacities of
women (MCSA, 2020). The Minerals Council intends to create a task force on women in mining
from among the Minerals Council and its member businesses, which will manage the white paper's
implementation as well as its monitoring and evaluation.
6. REFERENCE.
APUD, E. & MEYER, F. 2009. Ergonomics for the mining industry. Santiago, Chile: Ideo-grama. (Online) Available from:
https://journals.sagepub.com/doi/abs/10.1177/0018720812456701. (Accessed: 29/05/2021).
APUD, E. & MEYER, F. 2012. Ergonomics for workers Concepción, Chile: Trama. (Online) Available from: https://jour-
nals.sagepub.com/doi/abs/10.1177/0018720812456701. (Accessed: 29/05/2021).
ASHWORTH, G. MOLAPO, M. MOLEFE, N. SCHUTTE, S. & ZITHA, M. 2004. Human and social issues influencing
the incorporation of women into the mining workforce.(Online) Available from: www.coal-
tech.co.za/.../Task%202.8.1%20-%20Women%20in%20Mining.pdf. (Accessed: 30/06/2021).
BANERJEE, S., BANDYOPADHYAY, L., DASGUPTA, A., PAUL, B. & CHATTOPADHYAY, O. 2016. Work related
musculoskeletal morbidity among tailors: A cross sectional study in a Slum of Kolkata. Kathmandu Univ Med J
(KUMJ), 14(56), pp.305-10.
BANSAH, K.J. DUMAKOR-DUPEY, N.K. SAKYI-ADDO,G.B. 2017. Digging for survival: female participation in ar-
tisanal and small-scale mining in the Tarkwa mining district of Ghana. SME Annual Meeting: February.19–22.
BAXTER, R. 2019. 25 years of evolution: a view from the Minerals Council. Journal of the Southern African Institute of
Mining and Metallurgy, 119(8), pp.14-16.
BELLO, I. 2020. Women, Ergonomics and Repetitiveness. Safety and Health for Workers. Research and Practical
Perspective. IntechOpen. DOI: 10.5772/intechopen.82065. (Online) Available from: Women, Ergonomics and Repet-
itiveness | IntechOpen. (Accessed: 10/06/23).
BOTHA, D. & CRONJE, J.F. 2015. The physical ability of women in mining: Can they show muscle? Johannesburg, North-
west University. August. (Online) Available from: http://dx.doi.org/10.17159/2411-9717/2015/V115N8A1. (Ac-
cessed: 27/06/2021).
38
BOTHA, D. 2016. Women in mining: engaging men in the dialogue. Journal of Contemporary Management DHET accred-
ited ISSN 1815-7440 Volume 13. Pp 954-976.
BOTHA, D.2017. Barriers to career advancement of women in mining: a qualitative analysis. (Online). Available from:
https://hdl.handle.net/10520/EJC-6bf85c1be. (Accessed: 30/06/2021).
CARMODY, P.C., MATEO, J.C., BOWERS, D. AND MCCLOSKEY, M.J. 2017. Linguistic coordination as an unobtru-
sive, dynamic indicator of rapport, prosocial team processes, and performance in team communication. In Proceed-
ings of the Human Factors and Ergonomics Society Annual Meeting.Vol. 61. Sage CA: Los Angeles, CA: SAGE
Publications.
CHENG, Y. & KAWACHI, I. 2000. Association between psychosocial work characteristics and health functioning in
American women: prospective study. American Medical Women's Association, Vol. 320, 27 May, pages 1432-1436.
(Online) Available from: www.cdc.gov/niosh/womrisk.html. (Accessed: 10/06/23)
DE ARRUDA, A.F., AND GONTIJO, L.M. 2012. Application of ergonomics principles in underground mines through the
Occupational Safety and Health Management System--OSHMS OHSAS 18.001:2007. Work. 2012;41 Suppl 1:4460-
7. doi: 10.3233/WOR-2012-0119-4460. PMID: 22317408. (Online) Available from: Application of ergonomics
principles in underground mines through the Occupational Safety and Health Management System--OSHMS
OHSAS 18.001:2007 - PubMed (nih.gov). (Accessed: 03/05/2023).
DIAS, B. 2014, ‘Musculoskeletal disorders in the South African mining industry’, Doctor of Philosophy, University of the
Witwatersrand, Johannesburg.
DUARTE, J., MARQUES, A.T. & BAPTISA,S. J., 2021. Occupational accidents related to heavy machinery: a systematic
review. Safety, 7(1), p.21.
DUBE, G. & CHILUBA, B.C.2021. Ergonomic Factors Associated with Lower Back Pain Amongst Load-Haul-Dump
Truck Operators at Freda Rebecca Gold Mine, Bindura, Zimbabwe. JPRM 2021, 3(1): 50-59. doi:
10.21617/jprm2021.320.
EFTIMIE, A. HELLER, K. & STRONGMAN, J. 2009. Gender Dimensions of the Extractive Industries: Mining for
Equity. The World Bank.
HARMONY GOLD MINING COMPANY. 2008. Harmony sustainable development report. (Online) Available from:
www.har.co.za/files/Harmony. (Accessed: 26/06/2021).
HEINE, A. 2012. A model for managing the barriers of introducing women into a mining industry. March, 16. (Online).
Available from: http://hdl.handle.net/2263/28565. (Accessed: 23/07/2021).
JAMES, J.P. & SCOTT, P.A., 2009. Ergonomics in South Africa, and beyond the borders. In Ergonomics in Developing
Regions (pp. 361-366). CRC Press.
KABONGO, K.P., & NAIDOO, S. 2021. Factors associated with musculoskeletal pain in the past 12 months amongst female
miners in a South African gold mines. Department of Mineral Resources, Pretoria, South Africa. South African Journal of
Physiotherapy.
KALKIS, H., ROJA, Z., BABRIS, S. & REZEPINA, I. 2015. Ergonomics in Business–Approach for Organisation Effi-
ciency Improvement. SCIENTIFIC CONFERENCE ON ECONOMICS AND ENTREPRENEURSHIP SCEE’2015,
2015. 146.
KANDARE, M. 2017. Promoting Gender Equality in the Mining Industry. (Online). Available from: https://gender
links.org.za/news/promoting-gender-equality-in-the-mining-industry/. (Accessed: 10/07/ 2021).
KHAN, T. 2013. Mongolia: Raising Female Participation in the Large Scale Mining Sector. (Online) Available from:
https://openknowledge.worldbank.org/handle/10986/16499. (Accessed: 02/08/2021).
LAHIRI-DUTT, K. 2008. Digging to survive: women's livelihoods in South Asia's small mines and quarries. South Asia
Survey., 15 (2). 217-244
LEVINE, J. A. WEISELL, R. CHEVASSUS, S. MARTINES, C.D. BARBARA, B & COWARD, W.A. 2001. Women
under strain. Vol 294, Issue 5543, 812. October, 26.
MA, C.C., GU, J.K., CHARLES, L.E., ANDREW, M.E., DONG, R.G. & BURCHFIEL, C.M. 2018. Work-related upper
extremity musculoskeletal disorders in the United States: 2006, 2009, and 2014 National Health Interview Sur-
vey. Work, 60(4), pp.623-634.
MANGAROO-PILLAY, S. & BOTHA, D. 2020. An exploration of women’s workplace experiences in the South African
mining industry. The Southern African Institute of Mining and Metallurgy
MATSHINGANE, L.B. 2017. Experiences of Women Working Underground at a Coal Mine in Mpumalanga Prov-
ince (Doctoral dissertation).
MINE HEALTH AND SAFETY COUNCIL. 2017. Personal protective equipment for women in mining: All you need to know
about PPE for women in the mining sector. (Online) Available from: http://www.mhsc.org.za/sites/de-
fault/files/PPE%20For%20WIM%20 Handbook%20-%20English.pdf. [accessed 05 June 2023].
MINE MAGAZINE. 2019. Are women the mining industry’s most underdeveloped resource. (Online) Available from:
http://minesmagazine.com/8749/. (Accessed: 01/07/2021).
MINERAL COUNCEL OF SOUTH AFRICA. 2020. Women in Mining South Africa: Fact Sheet. March
NATIONAL INSTITUDE FOR OCCUPATIONAL SAFETY AND HEALTH (NIOSH). 2019. Mining topic: ergonom-
ics and MSD prevention. Center for Disease Control and Prevention.(South Africa, 1993)
PELDERS, J.J & RIDDER, H. 2020. "Assessment of the ergonomie design of self-contained self-rescuer (SCSR) devices for
use by women in mining." Journal of the Southern African Institute of Mining and Metallurgy 120.5 (2020): 307-312.
(Online) Available from: https://researchspace.csir.co.za/dspace/bitstream. (Accesses: 30/06/2021).
39
REPUBLIC OF SOUTH AFRICA. 2010. Amendment of the Broad-based Socioeconomic Empowerment Charter for the
South African Mining and Minerals Industry. (Online) Available from: http://www.dmr.gov.za/publications/sum-
mary/108-mining-charter-downloads/128-amendedofbbseecharter.html. (Accessed: 30/06/2021).
RICKARD, S. TREASURE, W. MCQUILKEN, J. MIHAYLOVA, A. & BAXTER, J. 2017.Women in Mining, Can a
Mining Law Unlock the Potential of Women? Adam Smith International and International Women in Mining, London.
SCHUTTE, P.C. 2010. Fatigue risk management: charting a path to a safer workplace. Journal of the Southern African
Institute of Mining and Metallurgy, vol. 110. no. 1. pp. 53-55.
SHARIAT, A., MANOHAR, A., SHAMSUL, B.T., AND MAHMOUD, D. 2016. Review musculoskeletal disorders and their
relationship with physical activities among office workers: a review. Malaysian Journal of Public Health Medicine 16(1):62-
74. January.
SOUTH AFRICA. 1996. Mine Health and Safety Act, Act No. 29 of 1996. Department of Minerals and Energy. Pretoria.
(Online). Available from: www.gov.za/documents/mine-health-and-safety-act. (Accessed 29/06/2021).
SOUTH AFRICA 1993. Occupational Health and Safety Act 83. In: LABOUR, D. O. (ed.). Pretoria: Government gazette.
SOUTH AFRICA. 2002. Mineral and Petroleum Resources Development Act, Act No. 28 of 2002. Department of Minerals
and Energy. Pretoria. (Online). Available from: https://www.gov.za/.../201409/a28-020.pdf. (Accessed: 29/06/2021)
SOUTH AFRICA. 2019. Ergonomic Regulations.
STEVEN, J. 2013. Research Paper Ergonomic Hazards Associated with Small Scale Mining in Southern Africa. Depart-
ment of Geography and Environmental Studies, Midlands State University, Gweru, Zimbabwe. International Journal
of Pure and Applied Sciences and Technology ISSN 2229 – 6107. 8-17. (Online). Available from: www.ijopaasat.in.
(Accessed from: 20/07/2021).
ZUNGU, L. 2011. Women in the South African mining industry: an occupational health and safety perspective. Inaugural
lecture, Unisa, Pretoria, October, 20. (Online) Available from: http://uir.unisa.ac.za/.../inaugurallec-
ture_Women%20in%20the%20SAMi_LIZungu_20October2011.pdf?. (Accessed: 01/ 10/22).
40
Is carbon offset realistic for South African mining industry

Zanele Jordaan
Latona Consulting Pty Ltd, Johannesburg, South Africa
Abstract
South Africa ranked fourteenth largest emitter of carbon dioxide worldwide in 2021. Fugitive emission of
carbon dioxide and methane from mining is considered as main sources of Greenhouse Gas emissions. The
mining industry is estimated to contribute about 7 percent to the total industrial process and product used (IPPU)
sector and greenhouse gas emission in this industry arise mainly from the energy acquired and used for the
operation, extraction and transportation of the commodity being mined (Department of Mineral Resources of
South Africa 2020). The South African transparency report denotes that the country needs to reduce its
emissions to below 348 MtCO2 by 2030 and to below 224 MtCO2e by 2050 to be within a 1.5°C ‘fair share’
pathway (South Africa Climate Transparency Report 2020). Carbon offsetting by reduction first then offsetting
the rest is a more productive approach for the better benefit of low emitters and the local communities around
the mines by optimizing carbon offset projects.
1. INTRODUCTION
The mining industry of South Africa contributes significantly to the economy and the revenue of a country, the
production of more than 50 different minerals from over 1700 mines. However, the recorded and measured data
of the greenhouse emission in the mining industry has been reported to be in the top 20 of the most carbon-
intensive global economies on emissions per gross domestic profit basis. The mining production is energy
intensive and due to that Climate change response and measures must be in place. The greenhouse gas emissions
in the mining industry contribute 7% to the total of the industrial process and product used sector (IPPU). A
substantial portion of the energy consumed and used in the mining sector comes from gas, coal, electricity and
petroleum products.
Greenhouse gas emission reduction strategies worldwide have been shared and encouraged as a way to assist
global climate change. In South Africa carbon tax was introduced in 2019 as a means to mitigate greenhouse
emissions in the country, the idea was that with the introduction of carbon tax the heavy emitters will move in
a direction of reducing their greenhouse emissions so as to lower their tax liability and then offset the Carbon
by investing in projects that will benefit and accommodate a net zero or carbon neutral economy which will in
turn support climate change. This is a desktop study research work, of which the interest was sparked by recent
work done on a gold mine sampling fugitive emissions on ventilation fans, previous work projects conducted
by Latona Consulting and personal communication with an international consultant specialising in Climate
Change and Carbon Offsetting strategies based in the UK on her recent visit in South Africa.
1.1. Greenhouse Gas Emissions

South Africa's greenhouse emissions between the years 1990-2017 have been recorded to have increased by
41%, However it has been published that the emissions in the recent years have declined due to new green
technology being introduced in the industry and the investments in verified carbon offset projects (Department
of Forestry, Fisheries & the Environment of South Africa 2022). Greenhouse gases such as carbon dioxide, methane
and nitrous oxide are the most common to refer to gases involved in climate change conditions. Primarily the
South African mining industry focuses on scope 1 emissions derived from the sources that the company or
organisation controls directly and scope 2 emissions which an organisation indirectly causes because of the
energy purchased that is needed to function is produced, generation of the electricity.
41
There are scope 3 emissions sources that the mining industry tends to side line as this is deemed an end user
emission segment. This is not produced by the company and the aftermath is not controlled by the company,
which is when the human source buys, utilise and dispose of products from suppliers. (National Grid 2023)
Figure 1: The main sources of scope 1, 2 and 3 emissions. (National Grid 2023)
The climate transparency report includes the notion that scope 1 and scope 2 emissions have to be reduced over
the next five to fifteen years with net zero emissions ideally by 2050. and in 2030 global CO2 emissions are
aimed to be cut by 45% below 2010 levels, this is a global collaboration 1.5 degree Celsius compatibility goal
to combat the adverse effects of for example, the extreme weather change as we have experienced in recent
years at a 1,1 degree Celsius global warming risk index (South Africa Climate Transparency Report 2020).
A national climate mitigation and adaption strategy has been in place and not particularly in effective motion.
The greenhouse gas emissions have since reached a stable state due to some reduction as technology improves
in the country even though scope 3 emissions are not easy to reduce as companies cannot control the end user
behaviour patterns in the use of the products, however, if the customer and organisation collaboration is initiated
whether by awareness of how their consumer can also be part of the solution in emission reduction goal that
can ease the carbon dioxide trend that has increased due to high demand of end user production. (National Grid
2023)
Figure 2: Energy related CO2 emissions by sector.
42
Figure 3: Gas contribution to South Africa’s emission (excl. FOLU) between 2000 and 2020 (Department of Forestry, Fisheries & the
Environment of South Africa 2022).
2. MINING INDUSTRY EMISSIONS
2.1 Gold Mining

A project to sample fugitive emissions on a gold mine ventilation system, this consisted of carbon dioxide
𝐶𝑂! and methane 𝐶𝐻# content sampling in the mines return air, The average yearly gas emissions are given in
tonnes and an extrapolated approach to calculate the amount of trees needed to offset 𝐶𝑂! for the gold mine.
The annual 𝐶𝑂! offset rate varies from 22 kg 𝐶𝑂! per tree to 32 kg per tree, so to make up for 1 tonne of carbon
dioxide 31 to 46 trees are needed. Calculation below is to give an extrapolated example for the operating gold
mines in South Africa, this is unpublished content from research submitted by Mani Dlamini in fulfilment of
an undergraduate Mining Engineering qualification at University of South Africa (UNISA).
For this operating gold mine the total estimated 𝐶𝑂! released it accounts for on average is about 15 834
tonnes per year and methane 𝐶𝐻# released about 130.6 tonnes per year.
15 834 + 130.6 ≈ 16 000 tonnes per year
To satisfy the notion of 1 tonne of 𝐶𝑂! 31 to 46 trees would be required.
½ (31 + 46) = 38.5 ≈ 39 trees
For this one gold mine number of trees needed to indemnify carbon offset 1 tonne of 𝐶𝑂! and 𝐶𝑂! e
16 000 𝑥 39 ≈ 624 000 trees
43
To extrapolate total emissions a calculation to accommodate 86 gold mines currently active in South Africa,
would have to be done, so we can say it would need 86 𝑥 624 000 trees. This is based on data available for this
project and other gold mines may have higher or lower emissions than what is measured and sampled. Also,
this project was a short term sampling. For more accurate calculations, long term sampling and data capturing
can be done, shared information amongst the ventilation and environmental department in the mine is crucial
for the success of the sustainability goal to be achieved.
2.2 Coal Mining

The coal mine project done by Latona Consulting for Coaltech, measuring total greenhouse emissions for
abandoned mines, surface and underground mines, fugitive emissions and discard dumps. The project
conducted in 2012 the total emissions in coal mines is calculated to between 70 000 to 80 000 tonnes a year of
𝐶𝑂! and CH4 to calculate, the same number of 39 trees as per the previous estimate will be deduced in this
estimate for coal mines.
For carbon offsetting 70 000 tonnes of emissions of 𝐶𝑂! and 𝐶𝐻# per year
70 000 x 39 = 2 730 000 trees
To estimate carbon offset for 80 000 tonnes of emissions of 𝐶𝑂! and 𝐶𝐻# per year
80 000 x 39 = 3 120 000 trees
The sampling conducted included that of the rehabilitation site of an open cast mine that had grass implantation
to offset the carbon, the outcome of the results showed that there was no 𝐶𝑂! found in the grassland, which
means that rehabilitation by grass for emission reduction and mitigation has been achieved in that site and it's
important that the open cast mines keep rehabilitation up to date and keep sampling to see if there are any
changes over time and rectify if need be.
3. DISCUSSION AND CONCLUSION
3.1 Climate Change

The IPCC assessment report released in 2021 confirms that climate changes will increase in the coming decades
if nothing is done to mitigate and reduce the current GHG emissions. For 1.5 and 2 degree Celsius global
warming, extreme heat temperatures will occur, longer warm seasons and shorter cold seasons. However,
climate change is not just about temperature changes but the effects this will have on agriculture, rising sea
level and dissipating Arctic Sea ice moreover the health effects to humans having to experience this in the future
(SAPPI 2023). It is a clear threat to humanity, the environment and future generations. Therefore, the mining
industry can play a significant role in the transition to a low carbon globe. A short term to long term planning
approach can be the most cost effective, low emitters can start early to being part of the solution thus making it
cheaper at that phase by starting with plantation of trees, note that this is a short term climate change mitigation
strategy dependent on the life of the trees and weather conditions that affect early afforestation (Oxford
University 2020). This is a carbon neutral approach not the net zero as mine would be prioritising to reduce its
own emissions at this stage.
44
3.2 Carbon Offsetting

The Carbon Offset Administration System of South Africa has 34 registered projects (Department of Mineral
Resources and Energy 2022). However, that number speaks to something other than the current active mines in
South Africa as of January 2023 which is over 500 mines. A carbon offset broadly refers to a reduction in
Greenhouse Gas emissions – or an increase in carbon storage. The most common methods of carbon offsets
such as planting a million trees every year, the current planting of grass in and around some mining areas, which
are more realistic and cost-effective as they are readily available in the country, but they are not sufficient to
maintain net zero in the long run since they are short lived carbon storage even though this has sustainable
benefit such as biodiversity conservation. The mining industry can invest in other natural climate solutions that
are identified as verified carbon offset projects that can directly improve the water and air quality of host
communities around the mines. (Oxford University 2020)
Much research can be done, along with long term emission sampling, data capturing, transparency and good change
management to identify new and innovative ways for carbon offsetting. The investment in especially local projects will
boost the socioeconomic state of the country, as it will create more jobs meaning poverty alleviation and better the
health of the people (Frumkin 2013). While the mine lowers its carbon tax liability which will in turn make
mining production costs efficient. The Minerals Council of South Africa encourages integrating renewable
energy into the mining operations and using smart energy optimisation systems.
Another storage solution for the mining industry is Carbon Mineralization, which is a chemical reaction that occurs
when CO2 turns into a solid material such as a carbonate in certain rocks. This process offers a permanent carbon capture
storage solution because the CO2 turns into stone and therefore cannot transmit back into the atmosphere. This
is by injection of CO2 into rock formations deep underground or exposing broken rock pieces at the surface of
mine tailings to CO2 (Carbon Credits 2022).
The verified carbon offset standards are allowed into the South African offset scheme in the Clean
Development Mechanism (CDM), Verified Carbon Standard (VCS) formerly known as Verra Standard and the
Gold Standard (GS) (Department of National Treasury of South Africa 2018). In South Africa, an example of
a carbon offset registered project that will directly benefit the host communities and the country at large is the
Wonderbag This is said to be a heat retention cooker that promotes better air quality in the home and energy
saver because the heat required is to only start the cooking process thereafter the Wonderbag continues cooking
as it has insulating properties that the bag retains heat with no further energy like electricity or biofuel. This is
manufactured by a South African social enterprise that also benefits job creation in the country.
Figure 4: Indoor air pollution from traditional cooking- source Google.
45
Figure 5: Design and demonstration of use of the Wonderbag.
The mining industry in collaboration with the government would have to perform a due diligence that will be
best designed in accordance with heavy emitters to low emitters categories as investment carbon offset projects
(G. Cook 2023). There is one goal in mind and everyone has to participate. Identify carbon offset projects that
are registered and comply with a South African verified standard. Creative and Innovative technology in
conjunction with renewable energy for achieving green mines goal. The projects must demonstrate
methodology to reduce emissions and have good likelihood of performance for ongoing reduction in emissions.
Change management needs to be set in place by the mine and preserve the integrity of correct emission sampling
and transparency in data so that there is clear representation of net zero to be achieved.
4. REFERENCES
Carbon Credits. 2022. “Generating Carbon Credits from Mining Waste.” Carbon Credits. 23 June.
carboncredits.com/bhp-carbon-mineralization.
Cook, A.P. 2013. Fugitive Greenhouse Gas Emissions from South African Coal Mines. Latona Consulting, Coaltech.
Cook, A.P. 2012. Gas emission from coal discard dump. Latona Consulting, Coaltech.
Cook, A.P. 2012. Gas Emissions from Surface Coal Mines. Latona Consulting, Coaltech.
Cook, A.P. 2012,1a1b. Gas Emissions from Surface fan emissions and Underground coal mines. Latona
Consulting, Coaltech.
Cook, A.P. 2012. Greenhouse Gas abandoned surface coal mines. Latona Consulting, Coaltech.
Cook, A.P, and P.J.D Lloyd. 2012. “The estimation of greenhouse gas emission from South African surface
and abandoned coal mines.” South African Institute of Mining and Metallurgy, online.
Cook, G.R, interview by Z Jordaan. 2023. Personal Communication - Carbon Offsetting Projects outside of
South Africa
Department of Forestry, Fisheries & the Environment of South Africa. 2022. “Consultation on the 8th National
Greenhouse Gas Inventory Report for the Republic of South Africa.” no. 2321. South Africa: DFFE, 29
July. 8thnational-greenhouse-gasinventory-reportg47133gon2321.
Department of Mineral Resources and Energy. 2022. “South African Carbon Offset Administration
Programme Draft.” Directorate Energy, July.
Department of Mineral Resources of South Africa. 2020. “Carbon Tax, It's Implications to the Coal Industry
and South African's Economy.” South Africa: Directorate Mineral Economics.
https://www.dmr.gov.za/Portals/0/Resources/Publications/Mineral Economics/Mineral Economic
Reports/R131-2020 Carbon Tax.pdf?ver=2021-06-01-104555-010.
Department of National Treasury of South Africa. 2018. “Explanatory Memorandum to the 2018 Carbon Tax
Bill - 20 Nov 2018.” South Africa: National Treasury of South Africa, 20 November.
https://www.treasury.gov.za/comm_media/press/2018/2018112101%20Explanatory%20Memorandum%20
to%20the%202018%20Carbon%20Tax%20Bill%20-%2020%20Nov%202018.pdf.
46
Dlamini, M. 2023. Emissions of Carbon Dioxide and Methane from South African Gold Mining (research
report for marking). University of South Africa, College of Science, Engineering and Technology,
Department of Mining Engineering.
Elston, L. 2021. “Why South Africa's carbon offset market is looking to expand.” Energy Monitor. 22 January.
https://www.energymonitor.ai/policy/carbon-markets/why-south-africas-carbon-offset-market-is-looking-
to-expand/.
Frumkin, Howard. 2013. “Environmental Health: From Global to Local.”
National Grid. 2023. “What are scope 1, 2 and 3 carbon emissions?” National Grid. 01 01.
https://www.nationalgrid.com/stories/energy-explained/what-are-scope-1-2-3-carbon-emissions.
Oxford University. 2020. “The Oxford Principles for Net Zero Aligned Carbon Offsetting.”
https://www.smithschool.ox.ac.uk/sites/default/files/2022-01/Oxford-Offsetting-Principles-2020.pdf.
SAPPI. 2023. “Climate Change Sustainability FAQ 11.” www.sappi.com.
Sasol Limited. 2022. “Climate Change Report 2022.” 30 June. https://www.sasol.com/sites/default/files/2022-
08/2022%20Sasol%20Climate%20Change%20Report_5.pdf.
South Africa Climate Transparency Report. 2020. “www.climate-transparency.org.”
47
The role of artificial intelligence in enhancing mine ventilation,

emergency preparedness, and occupational hygiene
FH Jacobs
AMS Haden
ABSTRACT: Mining operations are inherently hazardous environments, posing various risks and challenges
to worker safety, health, and productivity. Effective mine ventilation, underground emergency preparedness,
and occupational hygiene practices are crucial for ensuring a safe and healthy work environment. However,
traditional approaches to these areas often fall short in addressing the complex and dynamic conditions of min-
ing operations.
This paper explores the role of artificial intelligence (AI) in enhancing mine ventilation, underground emer-
gency preparedness, and occupational hygiene. By harnessing real-time monitoring, advanced analytics, and
decision-making capabilities, AI has the potential to revolutionize these domains, improving safety, efficiency,
and sustainability in mining operations.
The paper begins by examining the current state of mine ventilation, underground emergency preparedness, and
occupational hygiene, highlighting the limitations and challenges faced by traditional approaches. It then delves
into the potential benefits and opportunities presented by AI in these areas, including enhanced safety through
real-time hazard identification and mitigation, optimized ventilation systems, and predictive maintenance. Ad-
ditionally, AI can provide decision-making support to management and workers, improving overall operational
efficiency and productivity.
However, alongside the potential benefits, the paper also addresses the challenges and risks associated with AI
implementation, such as data quality, ethical considerations, regulatory compliance, technical limitations, and
cost and resource implications. It emphasizes the need for a multidisciplinary approach, collaboration, and the
development of ethical guidelines to ensure responsible and transparent use of AI-powered systems.
To facilitate successful development and implementation, the paper provides practical recommendations, in-
cluding the importance of collaboration and partnerships between mining professionals, AI experts, regulators,
and stakeholders. It emphasizes the need for ensuring data quality, addressing ethical considerations, complying
with regulations, fostering technical expertise, considering cost and resource implications, and continuously
evaluating and improving AI-powered systems and technologies.
1. INTRODUCTION
Mining operations play a vital role in supplying essential minerals and resources for various industries. How-
ever, these operations are inherently hazardous, posing risks and challenges to the safety and well-being of
workers. Effective mine ventilation, emergency preparedness, and occupational hygiene measures are crucial
for ensuring the health and safety of workers while maintaining productivity and efficiency in mining opera-
tions.
Traditionally, mine ventilation and occupational hygiene practices have relied on static, rule-based systems that
struggle to address the dynamic and complex conditions of mining environments. However, recent advance-
ments in artificial intelligence (AI) offer unprecedented opportunities to revolutionize these aspects of mining
operations. AI technologies can provide real-time monitoring, analysis, and decision-making capabilities that
48
could enhance safety, optimize ventilation systems, enable proactive emergency response, and improve occu-
pational hygiene practices.
This paper aims to explore the role of AI in enhancing mine ventilation, emergency preparedness, and occupa-
tional hygiene. It will assess the current state of these practices, identify the potential benefits and challenges
of AI integration, and provide recommendations for the development and implementation of AI-powered sys-
tems. By leveraging the capabilities of AI, the mining industry can significantly improve safety, efficiency, and
sustainability while addressing the limitations of traditional approaches.
The subsequent sections of this paper will delve into the current state of mine ventilation, emergency prepared-
ness, and occupational hygiene, highlighting the challenges and limitations of conventional methods. It will
then explore the potential applications and benefits of AI, including improved safety, increased efficiency, en-
hanced sustainability, predictive maintenance, and decision-making support. Additionally, the paper will ad-
dress the challenges and risks associated with AI implementation, such as data quality, ethical considerations,
regulatory compliance, technical limitations, and cost-effectiveness.
To ensure successful integration, the paper will provide practical recommendations for the development and
implementation of AI-powered systems and technologies in mine ventilation, emergency preparedness, and
occupational hygiene.
2. MINE VENTILATION ENHANCEMENT
2.1 Traditional approaches and limitations

Current mine ventilation systems rely on periodic manual inspections, which are time-consuming and may not
provide real-time information on air quality and ventilation performance. This can lead to undetected hazards
and exposure to harmful substances, jeopardizing worker safety and health. Furthermore, the design of ventila-
tion systems is often based on simplified models that may not capture the complex airflow patterns and varia-
tions in underground mines.
2.2 Potential applications of AI

AI-powered systems have shown great potential in enhancing mine ventilation by providing real-time monitor-
ing, analysis, and predictive capabilities. Through the integration of sensor networks with AI algorithms, these
systems enable continuous monitoring of crucial air quality parameters including temperature, humidity, gas
concentrations, and airflow rates. By analysing this data, AI models can effectively detect anomalies and iden-
tify potential hazards such as poor air quality, insufficient ventilation, or the presence of toxic gases.
Furthermore, AI can play a vital role in optimizing the performance of ventilation systems. By dynamically
adjusting fan speeds, damper positions, and air distribution based on real-time conditions and predictive models,
AI can ensure the efficient utilization of power and resources. For example, during periods of lower activity or
in specific areas with lower occupancy, AI algorithms can intelligently reduce ventilation rates, resulting in
significant energy savings without compromising safety.
Moreover, AI-powered ventilation systems can adapt to changing environmental conditions and proactively
respond to emerging risks. By continuously monitoring and analysing data, AI algorithms can identify potential
49
issues in advance, allowing for timely interventions and adjustments to maintain optimal airflow and air quality
throughout the mine.
Overall, the integration of AI in mine ventilation holds great promise in terms of improving air quality, opti-
mizing ventilation system performance, and achieving significant power savings, all of which contribute to a
safer and more efficient underground mining environment.
2.2.1 Data Quality and Quantity

In mine ventilation, the reliability and accuracy of sensor data are crucial for effective AI analysis. However,
several factors can affect the quality and quantity of sensor data. For instance, environmental conditions, such
as dust, humidity, and temperature fluctuations, can interfere with sensor accuracy, leading to erroneous read-
ings. Additionally, sensor failures or malfunctions can result in missing or corrupted data, which can hinder the
performance of AI algorithms.
2.2.2 Ethical Considerations

When implementing AI in mine ventilation, ethical considerations must be addressed to ensure responsible and
transparent use of the technology. Data privacy is a significant concern, as AI systems may collect and analyse
sensitive information about mine workers and operations. It is crucial to establish clear guidelines and protocols
to protect the privacy of individuals and comply with relevant data protection regulations. Moreover, transpar-
ency of AI algorithms is important to understand how decisions are made, detect potential biases, and gain trust
in the system.
2.2.3 Regulatory Compliance

Compliance with regulations and standards is essential to ensure the safe and responsible use of AI in mine
ventilation. However, integrating AI systems into existing ventilation infrastructure may pose challenges to
meet regulatory requirements. For example, ventilation standards and guidelines may specify specific parame-
ters, thresholds, or control strategies that need to be followed. If the AI system deviates from these requirements
or fails to comply with regulatory standards, it may raise concerns regarding worker safety, health, and legal
compliance. Therefore, thorough testing, validation, and close collaboration with regulatory bodies are neces-
sary to ensure that the AI-powered ventilation system conforms to relevant regulations and guidelines. Outdated
regulations can hinder the compliance of AI technology in mine ventilation. These regulations may not account
for the unique characteristics of AI systems or provide clear guidelines for their implementation. This misalign-
ment creates ambiguity and delays in ensuring compliance.
2.2.4 Technical Limitations

The harsh and challenging mining environment can introduce technical limitations when implementing AI in
mine ventilation. The integration of AI systems with existing ventilation infrastructure may require modifica-
tions or adaptations to accommodate the technology. For example, the placement and installation of sensors in
remote or inaccessible areas of the mine can be challenging. Figure 1 shows how information can be displayed.
Connectivity issues, power supply constraints, and system compatibility with legacy equipment are also tech-
nical considerations that need to be addressed. Additionally, the performance of AI algorithms can be affected
by the complexity and variability of mining operations.
50
Figure 8: A typical display of information visualized by an AI interface
3. EMERGENCY PREPAREDNESS
3.1 Current state and challenges
The current state of emergency preparedness in mining operations faces significant challenges and shortcom-
ings. One of the primary challenges is the complexity and unpredictability of mine environments. Underground
mines are characterized by a wide range of hazards and risks, including cave-ins, gas leaks, and fires. Antici-
pating and preparing for these emergencies can be challenging due to the unique nature of each mine and the
potential for unforeseen events.
Another challenge is the limited accessibility and communication within underground mines. The underground
environment restricts communication channels, making it difficult to quickly relay information and coordinate
response efforts during emergencies. This limitation can delay the mobilization of rescue teams and hinder
effective decision-making, putting the safety of workers at risk.
Furthermore, inadequate availability and reliability of emergency response equipment and systems pose signif-
icant challenges. Outdated or malfunctioning equipment can hinder rescue operations and compromise the ef-
fectiveness of emergency response. Additionally, insufficient training and preparedness of workers can impact
their ability to respond appropriately in emergency situations.
AI can improve emergency preparedness through real-time monitoring, early hazard detection, and predictive
modelling. Sensor networks and AI algorithms can continuously monitor environmental conditions, including
gas concentrations, temperature, seismic activity, and airflow patterns. AI models can detect abnormal condi-
tions, trigger alarms, and provide early warnings of potential emergencies. AI can also support evacuation plan-
ning by simulating different scenarios, identifying safe routes, and optimizing evacuation strategies. Real-time
situational awareness can be enhanced through the integration of sensor data, mine maps, and AI-powered
51
decision support systems (refer to figure 2). By integrating AI algorithms with the mine's lighting system, it
becomes possible to develop smart lighting solutions that provide visual guidance to employees in real time.
Figure 9:Real-time refuge chamber conditions display with visual display.
3.3 Challenges and risks
Implementing AI in emergency preparedness presents certain risks and challenges that need to be addressed.
These challenges include:
3.3.1 Data Availability
Underground mines may face limitations in terms of real-time sensor data availability. The harsh and complex
mining environment can hinder the establishment of reliable communication networks and the collection of
timely data. This limitation can impact the accuracy and effectiveness of AI algorithms, as they rely on up-to-
date and comprehensive data for analysis and decision-making.
3.3.2 Algorithm Robustness
AI algorithms used in emergency preparedness must be robust and adaptable to handle various emergency
scenarios and evolving conditions. Ensuring the algorithms' reliability and performance in different situations
is essential. Developing algorithms that can accurately detect anomalies, predict hazards, and provide timely
insights requires rigorous testing, validation, and continuous improvement.
52
3.3.3 Integration with Existing Systems
Integrating AI systems with existing emergency response infrastructure and communication networks is a
complex task. It requires careful planning and coordination to ensure seamless operation and compatibility
between AI systems and other components of the emergency response system. Integration challenges can
arise due to differences in data formats, communication protocols, and system architectures.
3.3.4 Regulatory Compliance
AI's impact on regulatory compliance in emergency preparedness introduces challenges. Existing regulations
may not explicitly address the use of AI technology, requiring interpretation and adaptation to ensure compli-
ance. Ensuring transparency and explainability of AI algorithms, addressing data privacy and security con-
cerns, and meeting ethical standards further contribute to the complexity of regulatory compliance.
3.3.5 Human-Machine Interaction
The effective integration of AI in emergency preparedness involves considering the interaction between hu-
mans and AI systems. Designing user-friendly interfaces, providing proper training and education to person-
nel, and fostering a collaborative environment between humans and AI systems are essential for successful
implementation. Overcoming resistance or reluctance to adopt AI technology and addressing concerns about
job displacement are also important aspects.
Figure 3: Different graph types that can be used to incorporate data
53
4. OCCUPATIONAL HYGIENE ENHANCEMENT
4.1 Current state and limitations

Occupational hygiene in mining involves managing worker exposure to hazardous substances and maintaining
a healthy working environment. Traditional approaches are often reactive and may not effectively prevent
health risks or optimize hygiene measures. Monitoring of exposure levels is typically done periodically or
through manual sampling, which may not capture real-time variations or adequately represent individual expo-
sures.

AI can enhance occupational hygiene in mining by providing real-time monitoring, early detection of exposure
risks, and data-driven recommendations for preventative measures. Sensor networks and AI algorithms can
continuously monitor airborne contaminants, noise levels, vibration, temperature, and humidity. AI models can
analyse this data, identify exposure patterns, and predict potential health risks. Figure 4 shows a typical example
of an exception report that is fit for purpose. AI can also support the development of personalized exposure
monitoring systems, providing workers with real-time feedback and warnings to minimize exposure. Addition-
ally, AI can optimize hygiene practices by recommending appropriate control measures, such as ventilation
adjustments, personal protective equipment (PPE) usage, and work scheduling.
54
Figure 4: Example of a personalised exception report that can be generated
55
4.3 Challenges and risks

The implementation of AI in occupational hygiene poses several challenges that need to be addressed to ensure
its successful integration. These challenges include:
4.4 Data Quality and Privacy

The accuracy and reliability of sensor data are paramount for effective AI analysis and risk assessment in oc-
cupational hygiene. Ensuring data quality through proper calibration, maintenance, and validation processes is
crucial. Additionally, data privacy concerns must be addressed to protect sensitive information. Secure data
storage and transmission protocols should be implemented to safeguard employee privacy and comply with
data protection regulations.
4.5 Algorithm Accuracy and Interpretability

AI algorithms used in occupational hygiene must exhibit high levels of accuracy to provide reliable insights
and predictions. The algorithms should be capable of handling diverse data sources and extracting meaningful
information. Moreover, interpretability is crucial to gain trust and acceptance from workers and regulatory bod-
ies. AI models should provide clear explanations and justifications for their decisions, enabling stakeholders to
understand and validate the algorithm's output.
4.6 Ethical Considerations

Ethical considerations play a significant role in the implementation of AI in occupational hygiene. Ensuring
fairness, transparency, and accountability in AI systems is essential. Biases in data collection or algorithmic
decision-making should be identified and mitigated to avoid discriminatory outcomes. Additionally, the ethical
use of AI technology, respect for worker autonomy, and the potential impact on employment should be carefully
evaluated and addressed.
4.7 Regulatory Compliance

Compliance with relevant regulations and standards related to occupational health and safety is crucial when
implementing AI in occupational hygiene. These regulations may govern aspects such as exposure limits, risk
assessment methodologies, and workplace safety practices. Ensuring that AI systems align with these regula-
tions and standards helps maintain a safe and compliant working environment.
4.8 Cost-Effectiveness
The cost-effectiveness of implementing AI in occupational hygiene should be carefully evaluated. This includes
considering the initial investment required for AI infrastructure, the ongoing maintenance costs, and the training
and re-skilling needs of employees. Assessing the return on investment and weighing it against the potential
benefits and improved occupational hygiene outcomes is necessary to make informed decisions.
56
Figure 5: Typical display of occupational hygiene stressors gathered from data
5. RECOMMENDATIONS
Based on the analysis of AI applications in mine ventilation, emergency preparedness, and occupational hy-
giene, the following recommendations are proposed:
• Foster collaboration and partnerships between mining professionals, AI experts, regulators, and stakeholders
to share knowledge and expertise.
• Develop robust and reliable sensor networks for data collection in mining environments, ensuring data quality
and accessibility.
• Address ethical considerations by implementing transparent AI algorithms, safeguarding data privacy, and
mitigating biases in decision-making.
• Ensure compliance with relevant safety regulations and standards when integrating AI systems into mining
operations.
• Conduct comprehensive cost-benefit analyses to evaluate the feasibility and economic viability of AI imple-
mentation in mine ventilation, emergency preparedness, and occupational hygiene.
• Provide training and education to mining personnel on AI technologies, promoting understanding and trust
in AI-powered systems.
6. CONCLUSION
The potential of artificial intelligence (AI) in enhancing mine ventilation, emergency preparedness, and occu-
pational hygiene is vast and promising. This paper examines the current state of these practices, identified the
limitations of traditional approaches, and explored the potential benefits and challenges of AI integration. By
leveraging AI technologies, the mining industry can significantly improve safety, efficiency, and sustainability
while addressing the complex and dynamic nature of mining environments.
Through real-time monitoring, analysis, and decision-making capabilities, AI-powered systems offer opportu-
nities to enhance worker safety, optimize ventilation systems, enable proactive emergency response, and im-
prove occupational hygiene practices. However, the implementation of AI in these areas also present challenges
57
that need to be carefully addressed, including data quality, ethical considerations, regulatory compliance, tech-
nical limitations, and cost-effectiveness.
To ensure successful integration and maximize the benefits of AI, it is recommended that mining professionals,
AI experts, regulators, and other stakeholders collaborate and form partnerships. Adequate investment in data
collection and management systems is crucial to ensure the availability of high-quality data. Ethical guidelines
and standards must be developed and implemented to address concerns such as data privacy, bias, and discrim-
ination. Compliance with regulatory requirements and guidelines should be prioritized, and technical expertise
should be fostered through training and education programs.
Consideration of cost and resource implications is essential to ensure the long-term sustainability of AI-powered
systems. Continuous evaluation and improvement processes should be implemented to monitor the effective-
ness and efficiency of these systems, while actively seeking feedback from workers and stakeholders.
By following these recommendations, the mining industry can harness the power of AI to create safer, healthier,
and more sustainable environments for workers. The integration of AI technologies in mine ventilation, emer-
gency preparedness, and occupational hygiene has the potential to revolutionize the industry, enhancing worker
safety, optimizing operations, and contributing to the overall productivity and success of mining operations.
In conclusion, embracing AI in these areas requires a multidisciplinary approach, collaboration, and responsible
implementation. By doing so, the mining industry can pave the way for a new era of advanced mine ventilation,
emergency preparedness, and occupational hygiene practices, setting higher standards for worker safety, oper-
ational efficiency, and environmental sustainability.
7. SUPPLEMETARY MATERIAL
1. Mining Safety and Health Research at NIOSH: Reviews of Research Programs of the National Institute for Occupational
Safety and Health (National Academies Press, 2007)
2. "Occupational health and safety in mining—status, new developments, and concerns" by R. M. Harrison, D. G.
Hicks, and T. E. Scobbie (Annals of Occupational Hygiene, 2009)
3. "Artificial Intelligence in Mine Ventilation" by W. L. Fourie and B. K. Belle (Proceedings of the 9th International
Mine Ventilation Congress, 2019)
4. "Artificial Intelligence in occupational hygiene: a review of Applications, opportunities, and Challenges" by T.
S. Bello and S. S. Dharmadhikari (AI & Society, 2020)
5. "Intelligent ventilation control system for underground mines based on IoT and big data" by Y. Chen, J. Xie, Y.
Yuan, and C. Wang (International Journal of Mining Science and Technology, 2018)
6. "AI in Mining: Mineral Exploration, Autonomous Drills, and More" by R. Faggiani (Emerj, 2020)
7. "Application of Artificial Intelligence in Ventilation Control of Underground Mine" by Z. Cui, H. Xia, and H.
Huang (IEEE Access, 2019)
8. "Review of ventilation systems optimization studies in underground mines" by N. Akbari Chermahini and A.
Kecojevic (International Journal of Mining Science and Technology, 2020)
9. "A Data-Driven Approach to Improve Occupational Health and Safety in Mining Using Machine Learning" by
A. Steiner, M. Nieto, and M. Ortiz (Resources, 2020)
10. "Artificial Intelligence Techniques in Occupational Safety and Health: A Systematic Literature Review" by I.
Gkiotsalitis, G. Sgouros, and N. S. Thomaidis (International Journal of Environmental Research and Public
Health, 2021)
11. "Machine Learning and Deep Learning Applications for Occupational Health and Safety: A Systematic Review"
by S. Wang, M. Singh, and D. Xue (International Journal of Environmental Research and Public Health, 2021)
58
12. "Integration of Artificial Intelligence for Occupational Safety and Health: A Review" by H. S. Lee, K. Kim, and
S. H. Han (International Journal of Environmental Research and Public Health, 2020)
13. "Real-time Ventilation Monitoring and Optimization in Underground Mines: A Review" by Z. Wu, L. Wang, and
G. Shi (Journal of Cleaner Production, 2019)
14. "Internet of Things in Underground Mine Monitoring and Safety: A Review" by X. Zhai, J. Li, and J. Xia (Sen-
sors, 2019)
15. "Artificial intelligence in mining: a framework for sustainable and responsible applications" by A. Hentschel, M.
Wagner, and M. Schachtner (Mineral Economics, 2020)
59
Novel ionic liquid additive for silica scrubbing
R. Bosch
Technical Director, Rubber Nano Products (Pty) Ltd
K.G. Schulz
Principal Electronic Engineer, Air Blow Fans.
ABSTRACT: The use of a novel ionic liquid as an additive in wet scrubbers has been shown to be effective at
improving float coal wet-scrubbing. The application and impact of this novel ionic liquid additive in quartz/sil-
icate dust types is now demonstrated and highlighted as compared to current methods of measuring dust col-
lection efficiency. The greatly enhanced capability of a novel ionic liquid additive for scrubbing silica is now
described.
1. INTRODUCTION
Respirable dust, with particle diameter ≤ 10 μm, has the capability to penetrate beyond the terminal bronchioles
and become deeply embedded in human lungs, leading to respiratory disease complications.
A novel ionic liquid additive has been tested and known to improve the wetting characteristics of water in the
application of wet scrubbing coal dust particulate matter. [1] The ionic liquid additive is inherently safe, and its
components are regarded as food-safe chemicals.
This trial will serve to demonstrate an increase in dust collection by using the ionic liquid additive at a modest
addition amount of 0.1% by volume in the water. The dust we are measuring is a graded silica material as
supplied by the mine.
There are many methods of measuring respirable dust concentration, and in presenting any achieved reduction
thereof, depending on the measurement equipment at hand. See for example the studies:
• (Gupta, Kumar and Schafrik, 2021), hereinafter referred to as GKS. [3]
• (Patts et al., 2016), hereinafter referred to as Patts. [4]
These data presentation differences make it difficult to fairly compare claimed results in a uniform and unbiased
manner. Consequently, in this analysis we shall present our results in a similar form as GKS, as their
experimental setup and measurement range of interest most closely resembles our own.
The safety of airborne dust has been correlated to the surface area of the dust particulate[5] and this is an
important principle that must be considered when analysing mass flow of dust particulate material. Small
particles are disproportionately light, since volume is cubic and the surface area is quadratic, the nanoparticle
aspect ratio is inverse linear. Smaller particles are thus in the majority responsible for toxicity and exposure
diseases.
In a South African context, there is a pressing need to revise the OEL for crystalline silica from 0,1 to 0,05 or
even to 0,025mg/m3 in the short term.[2]
This means a significant improvement in filtration efficiency is required to achieve these advised safety goals.
2. TESTING REGIME
2.1 CDC Wet Scrubber
A wet scrubber [Fanbox X1763] was attached to a F1061 Fan as diagrammed schematically below.
60
Figure 1: The CDC wet scrubber showing the locations of the PM sensors.
Air flow was verified to be symmetrical, with a measured Average Corrected Velocity of 23.4 m/s, and a
Volume Flow of 14,1 m3/s.
As airflow was symmetrical, sensors were placed on the centreline of the assembly, top dead centre, on the
outside of the assembly via pneumatic coupling tubes, as indicated above. This to restrict full submersion of the
sensor in run-off water.
Graded silica is supplied to the system upstream of the fan from a vibrating hopper. This is a non-constant mass
flow delivery system. The sensors are all recording data as a time series and some adjustments are required to
correlate data values to be contemporaneous.
Note the physical arrangement above, while not atypical, is different to the scrubber of GKS, as the orientation
of the filtration at an angle in that case will considerably change the aerodynamic properties of the scrubber
unit. Furthermore, they utilized much lower volumetric air flow characteristics compared to the CDC scrubber
utilized in this study. There is a difference in performance when the airspeed is lowered, and this is described
in the GKS paper. A full analysis of the scrubbing performance dependant on airspeed is beyond the scope of
this study. Here we are simply quantifying the improvement in scrubbing performance when adding ionic liquid
additive to the scrubber’s water feed.
2.2 Data that is measured.

The following data are reported by the SPS30 sensor:
1. Number concentrations in counts/cm3
2. Mass (Gravimetric) concentration in µg/m3
3. Typical particle size (average particle diameter) in µm.
These data streams are useful for our analysis and can also allow correlation to the safety limits of respirable
dust per unit time, as opposed to the usual method of reporting in mines. Historically, electronic real-time data
collection was impossible, and therefore gravimetric collection in cartridges over long periods of time were
used. This second data output enables easy direct comparison with historical data collection methods for the
same input.
61
A particularly useful data value that the SPS30 also produces is the typical particle size which is an indication
of mean particle size being measured in that sample interval. This output correlates with the weighted average
of the number concentration bins measured with a TSI 3330 optical particle sizer. This can be useful in detecting
large changes in particle distribution and is usually considered when we are looking for the start or end of dust
detection events.
Duration of measurement.
The duration of measurement is important only in so far as it meets the sampling constraints of the sensor being
utilized.
The scrubber system must of course be allowed to come to its dynamic equilibrium operating conditions so all
flows of air, flows of water, flows of additives (if any), and scrubber mesh wetting levels are stable.
Dust ingress into these systems is typically not constant in real-world wet scrubbing scenarios, and some of the
early generation optical sensors would easily be overloaded with particulate material and thus give measurement
and consequently analysis errors.
Due to the high sample rate of the SPS30, time-varying dust ingress is well tracked by the sensor, and thus
spurious dust loading events are well captured by the system, and thus such events are incorporated in the
analysis.
A further consideration to the duration of measurement is that this is a filtration process, which as in any energy
dissipative process, inherently will result in peak broadening of particulate events over time.
This phenomenon does mean we have to ensure that the tail of a previous dust event does not actually interfere
with the next phenomena being measured. This is of course what occurs in a continuous mass feed in that some
of the exit material being detected was introduced in a previous moment in time and the system dynamics are
holding the dust before letting it leave the filtration process. It will be demonstrable using a step increasing feed
and showing the duration of stabilization of the exit values for a certain setup.
While long period continuous assessment of data will eliminate errors of separate peaks and peak broadening,
it will still allow for hidden errors due to accumulation of short-term disturbances from dynamic equilibria of
the aerodynamic system at that mass feed of particulate. This is difficult to predict, and it is advised to test
performance across a range of stable mass feed amounts since reality will have variable amounts of dust
impinging the scrubber assembly.
3. DATA ANALYSIS PROCESS
The Sensirion detectors are collecting a range of data based on mass concentration and number count
concentration of particles of various sizes. These utilise real-time detection and the entire testing procedure is
then recorded with all the variable data for the different PM bins of interest (1, 2.5, 4 and 10µm).
The data is then read into suitable processing software (MatLab or Mathematica) and the performance of the
filtration is determined using Beta and Eta calculations.
The efficiency of a filtration system ηx can be calculated directly from the system Filtration Ratio (Beta Ratio,
βx). Now βx is defined as the ratio of number of particles upstream of a filter with diameter greater than x microns
(Ni,x), to the count of outgoing particles with diameter greater than x microns (No,x). This could equally be restated
as a ratio of count concentrations.
Eqn 1
Eqn 2
The Numerical Count Concentration in counts/cm3 of material entering the filtration system is measured at the
Neck Sensor, and the Exit sensor measures the material concentration leaving the filtration process.
62
This method is chosen by the Authors, but the work by Gupta, Kumar and Shafrik (GKS1) chose to utilize a
different optical sensor and reported their collection efficiency using the gravimetric (mass concentration)
information (µg/m3) [after having converted their own count concentrations to mass concentrations]. This
information was then represented as a collection efficiency calculated as in Eqn 3.
Eqn 3
This is done for different bins (i.e., using a de-normalized PDF1) and this collection efficiency was then shown
for the different PM bins of interest. This method of representing efficiency will now be referenced as GKS
collection efficiency. It is not the same mathematical definition as Eqn 1 and 2. As our sensor also directly
outputs mass concentration information, we will represent our data with the GKS methodology as well, to better
enable direct comparison between the two studies.
The reference source Patts et.al (Figure 2) used a cartridge collection system of total mass accumulated over
the day and then used a combustion method to determine the proportion of coal dust collected vs atmospheric
(non-combustible) dust. This is accurate but is not a real time data collection. The interesting point of their
method is it likely applies to the real collection capabilities of typical mining operations, and these values are
reported at different distances from the actual mining operation relative to the input air (as supplied by internal
ventilation). This study was not capable of comparing its efficiency directly due to the very different nature of
collection, but some correlation is possible since the distances that this study used should result in some
gravimetric dropping of material over that distance. What is interesting about the Patts et.al study is that it
determines both coal and silicate/rock type dust content and correlates that to the actual mining procedure and
distance mined. This is a very useful connection as will guide any safety considerations and conclusions made
by a mine operator as to the improvement of effect by using the additive.
1
Strictly speaking, GKS present an empirically measured Frequency Distribution fi = ni , where the frequency of an event fi is simply the count of
the number of times ni that the event occurred (i.e., a de-normalized PDF). This is easily converted to an empirical Probability Distribution Function
(PDF) by dividing each value by the total number of events N. We prefer the term de-normalized PDF instead of Frequency Distribution to prevent
any possible confusion with other analysis terms involving the word frequency.
63
Figure 2: Patts et al The collection of dust using cartridges at
Future work is expected to correlate the cartridge collection and valuation as per Patts et.al with co-located PM
sensors used in this study, to assist the mines in achieving better safety values.
3.1 The CDF of the silica material used in this study.

The cumulative density function CDF of the silica used in this trial is experimentally determined and the graded
silica was provided for this trial.
The region of importance for the wet scrubber is all the material that can be airborne which is typically material
<75µm in radius. This is typically recognized as the float coal size range and for consistency, the tests conducted
made use of this size range.
64
The inset graphic in figure 3 (shown enlarged in figure 4), shows the proportion of silica dust in the “float”
range of ≤ 75 µm. The CDF of the silica dust tested is done across a very large range of particle sizes (0.3 to
3000µm) and tests have only considered the float range for any pursuant calculations. This is the norm, as the
higher size and mass particles will not move in the airstream for any appreciable distance.
Figure 3: CDF of graded silica as supplied to CDC from Harmony 4.
Figure 4: CDF of "float" silica expressed from original CDF.
From figure 4, it can be seen that 11.5% of the silica sample is in the toxic respirable range (i.e., smaller than
10 microns) as expressed to the total silica in the complete CDF in figure 3.
65
3.2 Overview of all Neck/Exit Beta results using N concentrations
The additive is affecting several aspects of the basic mechanics of the filtration – droplet formation, wetting
and the performance of the mesh covered by a film of water / additive as a kinetic barrier.
Now recall that both GKS and Patts use collection efficiency as per Eqn 3 based on the mass (alternately mass
concentrations). Using the mass concentration data collected, the test produced associated Beta numbers as
per Eqn 2 to tabulate the mass-based Beta numbers for the experiments in Table 1 below.
Table 1: The Beta values of the system using Neck/Exit sensors and N concentrations
Test run Beta PM0.5 Beta PM1.0 Beta PM2.5 Beta PM4.0
Test 1 Water only - no silica, no additive 1.31541 1.31541 1.31541 1.31541
Test 2 additive, water and silica – short run 45.0154 51.4892 52.9077 53.2421
Test 3 Water and silica, no additive – mesh not cleaned 34.7501 42.8832 44.6601 45.0788
Test 4 additive, water and silica 95.0589 118.52 123.629 124.832
Test 5 Water and silica, no additive 23.364 29.3674 30.6646 30.9695
Test 6 Water and silica, no additive – Final-clean 9.82196 12.1899 12.7009 12.821
The test sequence above is for one continuous run of the scrubber with water, with or without silica entering
the system, and with or without additive in the water, to investigate various modes of operation.
Test 1 does not have any silica product moving through the scrubber, only water. The constant Beta numbers
of 1.31 present the removal of 23,7% of water entering the system by mechanical factors alone.
Tests 2, 3 and 4 in sequence present the build-up of steady state filtration.

While Test 2 showed sudden dramatic improvement in Beta when the additive was fed into the water feed, the
additive feed was purposefully removed before complete saturation of the mesh with the water/additive
mixture, in order to investigate system ramp-up dynamics. Consequently, during Test 2, peak Beta
performance had not yet been achieved.
In Test 3 the additive was switched off suddenly to investigate additive ramp-down dynamics. Beta numbers
declined for all particle sizes but for PM0.5. It is suspected that the time-dependence of ramp-down of Beta is
particle diameter sensitive, and will be fully explored in future.
Test 4 has the mesh screen completely saturated with water and ionic liquid additive, indicated long-term
steady-state Beta values.
Test 5 was again a ramp-down test of changes in Beta when additive was removed, but over longer time scales.
This is useful as beta ratio declining over time was tracked with water only flushing in the system. There is still
some residual ionic liquid additive impact at this point.
Test 6 is simply the long-term steady state Beta values continuing on from Test 5 (i.e., now all ionic liquid has
been fully flushed from the system, and the test is scrubbing with water only) .
Test 6 was chosen as the reported value for water-only scrubbing. Test 4 is chosen as reported value for the
ionic liquid additive.
3.3 Correlation to real mining data
3.3.1 The mining example from Patts et al.

The next step was to utilize the Patts paper to correlate the Beta values to their measured values. In their system
they measured an Eta of 92.5% on total respirable dust (coal and silicates) in the float range. The proportion of
their dust was 14% coal and 86 % silicate based as determined by combustion studies.
1. The mine produced around 45ug of dust per ton of coal mined. That was then scrubbed and had
a filtration ratio of Beta equal 13.33 (corresponding to Eta of 92.5%). The achieved exit dust was
thus 3.375ug/t of mined material. Experiments conducted during this trial with water only at PM4
66
reports Beta = 12.8, remarkably close to their 13.33, considering the different measurement methods
employed.
2. The maximum tonnage mined per shift of 8 hours is thus 29.63t. This implies that safety limit for
that mining operation is 88.88 ton of material for a full 3 shifts in 24 hours.
3. The data of mining output is thus less than that value and the mine is conforming to the dust
limitations.
3.3.2 Ionic liquid additive implications to the Patts system.
It is assumed that the ionic liquid additive behaviour will correlate to the scrubber used by Patts et al. Similar
geometry exists and similar water flow rates were used. During the testing regime a higher volumetric airflow
was used and was tested with dust impingement numbers far greater than the averages used in the Patts study
(real-world).
Total dust impingement in the respirable range in that system is less than 3600 μg total per day. The CDC
scrubber was handling and processing feeds of over 8000 µg/sec and achieving high value of filtration.
If the ionic liquid additive were introduced with its relative increase in Beta performance of 124.832 the
achieved exit dust would be 0.7825 μg/t mined.
This dust exit value is achieved by using the ionic liquid additive beta value for the 86% of the dust that is
siliceous and using the same Patts efficiency value for the 14% of the total dust that is coal (ignoring any
potential improvement that additive would bring). Table 2 expands on the calculations.
Table 2: Ionic liquid additive calculation of efficiency for Patts.
Mass flow in µg/t of coal Silica Dust Coal dust
Generated 38.7 µg/t 6.3 µg/t
Exiting scrubber 0.3100 µg/t (ionic liquid additive increased 0.4725 µg/t (at same Patts efficiency)
efficiency from mass flow determinations)
Total dust leaving scrubber 0.7825µg/t of mine output
The implications of this Beta value of ionic liquid additive is that the limit per 8-hour shift would be 127.79 ton
and thus total safe mining output of 383.387 ton per day of continuous operation.
This mine is averaging output of 60 to 78 ton per day currently which is slightly below the safety limit (88 ton)
as shown in Figure 7.
Figure 7: The mine output comparison using the Patts data and ionic liquid additive.
67
4. CONCLUSION
The ionic liquid additive will improve the filtration efficiency of a wet scrubber for siliceous dusts. The dosage
was low at 0.1% by mass in the normal flow of water. Other studies by the authors have shown potential to
reduce waterflow as well.
The tests in this study indicate that the ionic liquid additive effect is likely resident on the scrubber Mesh for
some time before it is washed away by the normal waterflow. This is shown by the decay characteristics of beta
performance over time.
The next sequence of tests will aim to show the continuous performance over many hours of operation and the
determination of what water flow rates are sufficient for effective operation. It was noted that much higher dust
impingement values were observed in the silica and ionic liquid additive tests and the system did not seemingly
suffer any degradation of performance. It will now be pertinent to find the overload limit of the current
configuration.
This will also assist in determining the Mesh performances and maximum mass flow rates that the ionic liquid
additive can handle in this Mesh configuration.
The above performance values of filtration will now also need to be scaled according to water flow rate to see
what level of water reduction can be achieved with the ionic liquid additive being used. This is an opportunity
to ensure that conditions remain safe, whilst consuming far less water in the scrubber (if sufficient droplet
formation is achieved).
These results are also indicative of what can be achieved for most siliceous materials – there are many minerals
that have some silicate content and ionic liquid additive is expected to be efficacious for wetting and having a
positive interaction with them. This influences a great percentage of the silicosis risks of many mining and
construction activities.
The ionic liquid additive has also been shown to be positively interactive with sulphate moieties and thus can
potentially by useful for polyhalite scrubbing as well.
The ionic liquid additive is improving the physical process of wetting and droplet formation which
improves the interaction of the dust material and the scrubbing solvent (water and ionic liquid additive
in this case). In conclusion - independent of geometry or spray methodology the ionic liquid additive will
offer improvement of collection efficiency or filtration efficiency.
5. REFERENCES
1. Bosch, R.M. and Schulz, K.G. (2022) ‘A novel dopant for improved safety and performance in wet scrubbing
of coal mining operations’, Journal of the Mine Ventilation Society of South Africa, 75:3, 29-36.
2. Department of Labour (2005) ‘National Programme for the Elimination of Silicosis’. Available at:
https://www.ilo.org/safework/info/public/WCMS_118112/lang--en/index.htm.
3. Gupta, N., Kumar, A.R. and Schafrik, S. (2021) ‘Laboratory Determination of Coal Dust Cleaning Efficacy
of a Fibrous Filter for Flooded-Bed Dust Scrubber’, Minerals, 11(3), p. 295. Available at:
https://doi.org/10.3390/min11030295.
4. Patts, J.R. et al. (2016) ‘Reducing float coal dust: Field evaluation of an inline auxiliary fan scrubber’,
MinEng. 2016 December ; 68(12): 63–68. doi:10.19150/me.6883. [Preprint]. Available at:
https://doi.org/doi:10.19150/me.6883.
5. Schmid, O. and Stoeger, T. (2016) ‘Surface area is the biologically most effective dose metric for acute
nanoparticle toxicity in the lung’, Journal of Aerosol Science, Volume 99, 2016, Pages 133-143, ISSN 0021-
8502, https://doi.org/10.1016/j.jaerosci.2015.12.006.
68
A review of the application of Continuous Real-Time Monitoring in

Noise Risk Management
K Ramoshaba
Minerals Council of South Africa, MOSH Learning Hub – Noise
ABSTRACT: In the year 2020, the Chief Executive Officer (CEO) Zero Harm Forum (CEOZHF) reviewed the
slow progress towards achieving the 2024 Occupational Health Milestones and agreed that there was a need to
improve the current measures in place to reduce occupational diseases and fatalities. Upon conclusion of en-
gagements with the Group Environmental Engineers, it was recommended that key occupational health indica-
tors must be adopted to track industry progress on the elimination of workplace exposures, and this could be
done using Continuous Real Time Monitoring (CRTM). CRTM serves as an early indicator for potential expo-
sure as it has the ability to provide exposure data with data logging, providing an exposure profile over a certain
period of time. The CRTM should be ideally linked to the mine’s telemetry/PLC system and key performance
indicators displayed in the mine’s control rooms to monitor compliance. Where over-exposures or control fail-
ures occur, the risk owner must take action to address the deviations. The article presents the application of
CRTM in Noise Risk Management through the review of previous research and the key considerations in CRTM
application together with the principles of Critical Control Management (CCM).
1. INTRODUCTION
1.1 Background
The South African Mining Industry (SAMI) is the country’s largest industry, having employed more than 400
thousand people and contributed R73.6 billion in taxes and contributed R493.8 billion to the Gross Domestic
Product in 2022 (Minerals Council South Africa, 2023). The processes associated with mining generate signif-
icant noise relating to drilling, blasting, crushing, and processing of the ore. Fink (2019) defines Noise as an
unwanted and/or harmful sound. Excessive occupational noise levels have been documented to cause Occupa-
tional Noise Induced Hearing Loss (NIHL), limiting employees’ ability to communicate and hear warning safety
signals as well as their general safety and productivity (Edwards et al., 2011). The effects of NIHL on employees
are long-term and have a significant impact on the quality of life long after they have left the mines. A recent
study by Ntlhakana et al. (2020), reported that the average noise exposure levels in the SAMI range between
63.0 and 113.5 dB(A), with almost 73% of the mining population being exposed to more than the Occupational
Exposure Limit of 85dB(A).
The SAMI tripartite stakeholders recommitted to achieving Zero Harm in the mining industry by committing
to the following regarding NIHL:
• By December 2024, the total operational or process noise emitted by any equipment must not exceed a
milestone sound pressure level of 107 dB(A) and
• By December 2016, no employee's Standard Threshold Shift (STS) will exceed 25 dB from the base-
line when averaged at 2000, 3000 and 4000 Hz in one or both ears.
In 2020, the Chief Executive Officer (CEO) Zero Harm Forum (CEOZHF) reviewed the slow progress towards
achieving the 2024 Occupational Health Milestones and agreed that there was a need to improve the current
measures in place to reduce occupational diseases and fatalities. Upon conclusion of engagements with the
69
Group Environmental Engineers, it was recommended that key occupational health indicators must be adopted
to track industry progress on the elimination of workplace exposures, and this could be done using Continuous
Real Time Monitoring (CRTM). CRTM serves as an early indicator for potential exposure as it has the ability
to provide exposure data with data logging, providing an exposure profile over a certain period of time. The
CRTM should be ideally linked to the mine’s telemetry/PLC system and key performance indicators displayed
in the mine’s control rooms to monitor compliance. Where over-exposures or control failures occur, the risk
owner must take action to address the deviations.
To ensure effective and efficient application of CRTM, the Minerals Council South Africa’s Mining Occupa-
tional Safety and Health (MOSH) Learning Hub takes the International Council for Mining and Metal
(ICMM)’s Critical Control Management into consideration when reviewing the application of CRTM in Noise
Risk Management in the workplace.
2. CRITICAL CONTROL MANAGEMENT APPLICATION IN CRTM
ICMM defines Critical Control Management as “an integral part of risk management that focuses on identifying
and managing the controls that are critical to preventing catastrophic or fatal events. Critical controls can either
prevent a serious incident from happening in the first place or minimise the consequences if a serious accident
was to occur (ICMM, 2015).” Critical Control Management consists of nine steps, six of which are required to
plan the Critical Control Management program before implementation in the last three steps (Figure 1).
Figure 10. Critical Control Management Process
70
CRTM highlights causation factors behind failed controls and high exposures and provides an opportunity to
promptly react to exposures and remedy controls to their original status. The application of CRTM also effec-
tively focuses on profiling high-risk activities and implementing critical controls that will reduce the risk, re-
sulting in improved outcomes for employees. The application of the Critical Controls in CRTM can be done by
subjecting the operation’s overall noise risk to the nine-step process of Critical Control Management and assess
the applicability of CRTM (Figure 2). The process detailed in Figure 2 can only be applied once the Critical
Control Management process in Figure 1 has been completed.
Figure 11. CRTM Application Process
Trigger Action Response Plan (TARP) Implementation (Step 4) is a process whereby the organisation defines
its own trigger levels and reaction required. This is based on the organisation’s Noise Risk Management strate-
gies and can be different for working places based on the activities and processes taking place. For example, a
normal working environment can be defined as any workplace with noise measurements below 82 dB(A) which
can be flagged as a green requiring no action, with a high-risk working environment defined as noise measure-
ments equal or above 104dB(A) which can be flagged red requiring the stop of all activities until the noise has
been reduced.
3. NOISE RISK MANAGEMENT
3.1 Understanding the risk
Manwar & Pal (2022) reported that the effective management of occupational health hazards needs to be iden-
tified in mining, with the purpose of eliminating or minimising exposure at source such that it does not pose
harm to employees’ health, productivity, and quality of life. This can only be achieved when all noise sources
have been identified and their characteristics are fully understood.
The mining industry’s noise management system is heavily reliant on noise measurements and dosimetry, and
it is envisaged that the understanding of noise sources will aid in an effective Noise Risk Management Strategy.
There is also a need to manage noise using the latest tools and techniques available, such as noise mapping to
71
better understand the noise risk. Detailed noise mapping can be used to identify the areas where the noise levels
exceed the legal limits and enable the design and implementation of preventive control measures (Manwar &
Pal, 2022)
3.2 Review of Noise CRTM Practices across the world
A study conducted by Marques & Pitarma (2020) on “Real-Time Noise Monitoring System Based on Internet
of Things (IOT) for Enhanced Acoustic Comfort and Occupational Health”, investigated the use of iSoundIoT
which incorporates open-source technology, while taking into consideration the sensitivity, flexibility and ac-
curacy of noise measurements in real-time for effective acoustic comfort. The solution provided a definitive
software solution for a practical, intuitive, fast access analysis of building noise pollution and trigger alerts in
the case of poor acoustic comfort to alert those in the building as well the responsible persons in order to address
the factors relating to poor acoustic comfort. Figure 3 illustrates the system’s architecture (Marques & Pitarma,
2020).
Figure 12. iSoundIoT Architecture
A research article by Sundaram et al., (2021) on the “Development of Real-time IoT based Air and Noise Mon-
itoring System” focused on addressing the increased noise pollution in Malaysia utilizing real-time noise mon-
itoring systems. The noise pollution monitoring system IoT technology, which uses an LM393 sound sensor to
measure noise levels, then sent to the ThingsSpeak server and graphically display the measurements in real-
time. The system also has a warning signal framework that notifies the people staying the area when they are
exposed to high levels of noise. The system’s architecture is displayed in Figure 4.
72
Figure 13. ThingsSpeak Architecture
Fernandez-Prieto et al. (2019) presented a novel approach to developing an autonomous system for environ-
mental noise monitoring. The system developed was highly scalable, easy to use, low-cost and low powered to
aid with the widespread adoption by various industries. Power usage was kept low by using Bluetooth Low
Energy communication for streaming the environmental noise. The detected environmental noise was integrated
into a system that could classify the measured noise sources, which used Mel-frequency cepstrum coefficients
to produce features of the noise and model Gaussian mixture models to classify the source of the noise. The
basic operational principle of the system is that the information collected from the noise sensors is sent to a
cloud storage service in real-time which enables visualization of noise levels, noise sources and battery level of
the sensors in a web application. Figure 5 illustrates the output information from the system in a noise map.
Similar results were obtained from a study conducted by Jezdović et al. (2021) on a crowdsensing platform for
real-time monitoring and analysis of noise pollution in smart cities.
Figure 14.User Interface of the web application
73
4. CRTM CONSIDERATIONS
The CEOZHF circular No.6-22 on the Management of Significant Occupational Health Hazards, indicates that
the following must be taken into consideration when applying CRTM in Noise Risk Management.
4.1 Trigger Action Levels (TAL)
When implementing CRTM, it is important to define TALs based on the operations’ noise risk assessment out-
comes and should be aligned to the list below:
• TAL-1: The level of the noise hazard and/or control performance parameters, as monitored by the
CRTM, for the initiation of primary response purposes.
• TAL-2: The level of the noise hazard and/or control performance parameters, as monitored by the
CRTM, for the initiation of secondary response purposes.
• TAL-3: Is the level of the noise hazard and/or control performance parameters, as monitored by the
CRTM, for the initiation of tertiary response purposes.
4.2 Warning Signal
A signal that should be initiated when the noise level ≥ 85 dB(A) < 104 dB(A). The risk owner of such a
workplace should inform the employees and ensure that the required controls have been adhered too.
4.3 Intervention Signal
A signal that should be initiated when the noise level ≥ 104 dB(A). The risk owner of such a workplace should
investigate that incident, implement corrective actions to reduce the noise level to below 104 dB(A).
4.4 Critical Control Management
Develop a Critical Control Management program that details a practical method of improving managerial
control over rare but potentially catastrophic events by focusing on the critical controls.
5. CONCLUSION
There has been a lot of application of CRTM for noise monitoring in other countries, with limited work in the
mining industry. The research presented in the article highlights the practicability of applying CRTM in Noise
Risk Management. The systems applied utilise IoT, which is used in the mining industry in most risk manage-
ment programmes.
The SAMI has also started working on the application of CRTM for noise monitoring, though the application
in Critical Control Management is very limited. Some operations are more advanced than others, and the MOSH
Noise Team documents these practices in the form of case studies and leading practices to share with the mining
industry for the purpose of adoption.
74
To reduce the risk of NIHL and improving the general employees’ health and safety, it is important for the
SAMI to utilise the latest available technology and software that enables proactive response to noise risks. This
will also assist in understanding the noise risk as well as the identification of effective critical controls to reduce
the exposure at source.
6. REFERENCES
1. Badruddin, M.B. et al. (2020) ‘IOT based Noise Monitoring System (NOMOS)’, IOP Conference Series: Mate-
rials Science and Engineering, 884(1), p. 012080. doi:10.1088/1757-899x/884/1/012080.
2. Fernandez-Prieto, J.-A., Cañada-Bago, J. and Gadeo-Martos, M.-A. (2019) ‘Wireless acoustic sensor nodes for
noise monitoring in the city of Linares (jaén)’, Sensors, 20(1), p. 124. doi:10.3390/s20010124.
3. Fink, D. (2019) ‘A new definition of noise: Noise is unwanted and/or harmful sound. noise is the new
“secondhand smoke”.’, Proceedings of Meetings on Acoustics [Preprint]. doi:10.1121/2.0001186.
4. ICMM (2015) Critical Control Management: Good practice guide, Open graph. Available at:
https://www.icmm.com/en-gb/guidance/health-safety/2015/ccm-good-practice-guide#:~:text=Critical%20con-
trol%20management%20is%20an,incident%20or%20mitigate%20its%20impact. (Accessed: 16 May 2023).
5. Jezdović, I. et al. (2021) ‘A crowdsensing platform for real-time monitoring and analysis of noise pollution in
Smart Cities’, Sustainable Computing: Informatics and Systems, 31, p. 100588. doi:10.1016/j.sus-
com.2021.100588.
6. Manwar, V. and Pal, A.K. (2022) ‘Noise mapping: An effective tool for monitoring noise-induced health haz-
ards in active mines’, Springer Geology, pp. 215–236. doi:10.1007/978-3-030-99495-2_9.
7. Marques, G. and Pitarma, R. (2020) ‘A real-time noise monitoring system based on internet of things for En-
hanced Acoustic Comfort and occupational health’, IEEE Access, 8, pp. 139741–139755. doi:10.1109/ac-
cess.2020.3012919.
8. Minerals Council South Africa (2023) Integrated Annual Review 2022. rep. Rosebank, Johannesburg: MCSA.
9. Moroe, N.F. et al. (2020) ‘Feedback-based Noise Management Matrix in action’, South African Journal of
Communication Disorders, 67(2). doi:10.4102/sajcd.v67i2.678.
10. Ntlhakana, L., Nelson, G. and Khoza-Shangase, K. (2020) ‘Estimating miners at risk for occupational noise-
induced hearing loss: A review of data from a south african platinum mine’, South African Journal of Commu-
nication Disorders, 67(2). doi:10.4102/sajcd.v67i2.677.
11. Sundaram, D. et al. (2021) ‘Development of real-time IOT based air and Noise Monitoring System’, Alinteri
Journal of Agriculture Sciences, 36(1), pp. 500–506. doi:10.47059/alinteri/v36i1/ajas21071.
75
Establishing a process for the setting of real-time detection system

alarms
R.E. Henderson, BS, MBA

GfG Instrumentation, Inc. Ann Arbor, Michigan, USA
ABSTRACT: In March 2023 the American Industrial Hygiene Association (AIHA) published a White Paper
that proposes a formalized a “standard of care” process for selecting a sensor system that is fit for purpose,
configured properly for detecting a condition, and announces that condition through signals that are most com-
monly considered and called alarms. The process is applicable to all types of real-time detection sensors and
systems, including those used to measure gases and vapors, particulates, heat stress, noise, and other workplace
conditions. This paper will present and explain the five step White Paper process and discuss its application
using practical examples of real time measurement.
1. INTRODUCTION
Need for a formalized process for establishing fitness for purpose and setting direct reading Real Time De-
tection System (RTDS) alarms.
With the advancement of detection technologies and miniaturization of components, real-time detection of the
hazards occupying interest of the safety and hygiene practitioner have expanded. Those same techno-logical
advancements now allow linking sensing and detection system data to automatically mitigate the con-ditions
(ventilation controls) or alert people to immediate action (fire alarm strobes, audible alarms, and per-sonal
safety instruments). Initial action is most frequently taken on the basis of instantaneous readings rather than on
the basis of statutory time history exposure limit calculations (e.g., STEL and TWA). Real-time in-struments
generally include STEL and TWA alarm calculations, but in most cases, the initial alarm is triggered by ex-
ceeding a real-time concentration limit. Unfortunately, there remains a lack of guidance for establishing a “fit
for purpose” process for establishing and setting real-time alarms when the Occupational Exposure Limit (OEL)
does not provide direct guidance.
The standard of care establishes five steps:

Step 1 – Statement of Purpose in target of condition to detect.
Step 2 – Selection of Objectives (parameters) around the condition to detect.
Step 3 – Selection of the device and determination that it is fit for purpose.
Step 4 – Establishment of an Alarm Set Point and setting expectations in response.
Step 5 – Communication of the results and determination.
The White Paper is intended to frame a common process that reasonably covers configuring appropriate alarms
on various types of real-time detections systems. In doing so, the process should sequentially assist with select-
ing the sensor and configuring the real-time system, including, but not limited to, its datalogging function and
alarm setpoint(s). The entire process, when correctly applied, will ensure the user correctly inter-prets the con-
dition(s) induced when an alarm is triggered and takes an associated action.
76
2. PRACTICAL EXAMPLES OF ALARM SETTING GUIDANCE IN ACTION
Setting alarms for direct-reading instruments (DRI) can be the most complex part of implementing a real-time
detection system (RTDS). Recent developments in exposure limit changes, emphasis on systematic review of
the capabilities of DRIs, advancements in sensor technology, and reduction in instrumentation cost have
prompted discussion in the occupational and environmental health and safety professional community regarding
how alarms set points or thresholds are determined. Factors such as increased adoption of statistical analysis
methods and changes in the social relationship between workers and employers have also contributed to the
iscussion of alarm setting process transparency.
New guidance available from AIHA (Establishing a Process for the Setting of Real-Time Detection System
Alarms, 2023) provides practical guidance for setting alarms. This process includes five (5) key steps to deter-
mine the appropriate alarm settings based on a systematic analysis of the signalling context. Each step is in-
tendedto functionally align multiple components, including RTDS fitness for purpose, systematic review of the
DRI, and stakeholder expectations to result in a system configuration that meets the needs the party designing
and deploying the RTDS in a specific exposure context.
Key to this process is the fundamental principle that each alarm must trigger a defined response – from a human,
mechanical, or information system. The purpose of an alarm is to prompt action as defined by the scenario: any
alarm that is not intended to result in an action only contributes to habituating recipients to alarm conditions.
Practical examples often help solidify the approach recommended by the guidance. The White Paper pre-sents
an example application for chlorine dioxide monitoring in a pulp mill. Additional abbreviated examples using
the five-step process are presented to highlight the practical applications of alarm setting using addition-al hy-
pothetical scenarios.
3. CASE STUDY: USE OF REAL TIME PERSONAL MONITORS WITH DIRECT READING NO2
SEN-SORS IN UNDERGROUND WELDING OPERATIONS
The USA Department of Energy operates a secure nuclear waste repository in the state of Nevada. The site is a
large-scale underground project excavated into a salt dome. The geology gives rise to the potential presence of
naturally occurring sulfides including hydrogen sulfide (H2S) and sulfur dioxide (SO2). Carbon monoxide (CO)
and nitrogen oxides (NOx) are potential contaminants as a function of heavy equipment exhaust. Other hazards
include the potential presence of combustible gas and oxygen deficiency. Portable real-time direct reading in-
struments (DRIs) are routinely used throughout the site to measure these hazards and ensure that toxic exposure
limits are not exceeded. Inert gas welding on stainless steel is a common procedure at the site.
Problem: The DRIs in use at the site are subject unstable readings, and frequent, suspected false alarms due to
suspected cross interference of contaminants on the sensors, and the alarm settings currently in use at the site.
Step 1: Establish statement of purpose.
Inert gas welding on stainless steel is known to produce oxides of nitrogen (NOx). Nitrogen dioxide (NO2) is
the most persistent and has the lowest toxic exposure limit. DRIs used by workers in the welding ar-ea are
needed to document that the 8-hour time weighted average (TWA) and 15-minute short term exposure limit
(STEL) toxic exposure limits for any contaminants that may be potentially present are not exceeded. Peak
alarms based on instantaneous readings are used to alert workers to stop operations and leave the affect-ed area.
77
Datalogging functions are required to provide documentation that exposure limits are not exceeded, as well as
trending analysis of such alarm conditions to determine supplemental controls (such as behavior modifications
of the work, or supplemental engineering considerations).
Step 2: Selection of objectives.
The primary exposure concern is with nitrogen dioxide (NO2) due to a very low occupational exposure limit
(OEL) utilized by the company. NO2 is a Designation B (Health Monitoring) hazard based on this OEL as-
signment. The objective of monitoring is warning the worker to move away when nitrogen dioxide exposures
exceed either the real-time peak, TWA, or STEL exposure limit.
Step 3: Selection of devices and determination of fitness for purpose.
The conditions at the site, the context of the work environment, and the toxicological basis for the Occupational
Exposure Limit (OEL) are considered in this step. NO2 sensors are installed in the mine’s existing four-gas,
five-gas, and single gas monitors. The available NO2 sensors are limited to those which are certified for use in
the DRIs. The NO2 sensors offer industry standard performance with regards to range and resolution of 0.1
ppm resolution over a measurement range of 0 – 50 ppm. The mine was unable to identify products with sig-
nificantly better performance than their currently implemented devices, which are made by a well-known man-
ufacturer.
A significant issue is the response of the NO2 sensor to interfering contaminants such as SO2 (which can cause
negative readings), as well as the effects of humidity on sensor performance. The AIHA Standardized Equip-
ment Specification Sheet (SESS) was used to help clarify these sensor performance issues. It was found that
the NO2 sensor self-adjusts to changes in ambient humidity, that CO and H2S do not affect the NO2 sensor.
While SO2 is potentially present at the site, it is not generally present at the same time as the NO2 pro-duced
by welding operations.
The facility operates continuously. The work-shift is typically 12 hours. TWA alarms are calculated on either
the projected, or the most recent elapsed 8-hour period. Based on historical exposure data, most emissions are
below the 8-hour TWA and STEL exposure limit values, with rare instances of exceedance.
The DOE facility in this example currently uses the 1994 American Conference of Governmental Hygienists
(ACGIH) Threshold Limit Value (TLV) for NO2 (1 ppm TWA and 3 ppm STEL). The facility seeks to follow
the most current ACGIH OEL of 0.2 ppm TWA. ACGIH has not established an NO2 STEL.
Historically, instantaneous “LOW” alarms for toxic contaminants were often set at the TWA exposure limit
(TWAEL). This is possible as long as the sensor limit of detection is above the adjusted OEL. It is not possible
if the exposure limit is less than the lower detection limit of the sensor, or if there are not sufficient incremental
steps between the Limit of Detection (LOD) and the alarm setting to avoid false alarms. As a result of techno-
logical limitations of the instruments at the site, the 1.0 ppm TLV-TWAEL for NO2 is too low to be used as
the alarm setting for the instantaneous LOW alarm.
78
Step 4: Establishing alarm setpoint and response process.
Through collaborative discussions, site management arrived at alarm value policies using the ACGIH guidance
that worker exposure levels may exceed 3 times the value of the TLV-TWA level for no more than 15-minutes
at a time, on no more than 4 occasions spaced 1 hour apart during a workday, and under no circumstances
should they exceed 5 times the value of the TLV-TWA level when measured as a 15-min TWA. In addition,
the 8-hour TWA is not to be exceeded for an 8-hour work period. Accordingly, the following alarm settings for
NO2 were specified for use at the site:
Low alarm: 3.0 ppm

High alarm: 5.0 ppm
STEL alarm: 3.0 ppm
TWA alarm: 1.0 ppm
Step 5: Summarizing the analysis for effective stakeholder communication.
A summary report was prepared and delivered to the client mining company, describing the results of the anal-
ysis for fitness for purpose, what the instrument could accomplish across the four agents of concern, (NO2,
SO2, CO and H2S) and how the alarm set points and specific defined employee alarm responses re-quired by
the workers were determined.
The report was reviewed and distributed to the business line and division management as well as the facility
superintendent. It formed the basis for a briefing of the OEHS staff, subcontractors, and operational staff. This
report provided key transparency for workers seeking information on their protection during dis-semination of
instructions on how to respond to alarms.
The analysis summary was also pre-emptively provided to the site regulator as evidence of ethical and compre-
hensive examination of the topic given specific regulatory, technology, and field conditions that set the context.
79
REFERENCES
1. AIHA. (2021a). About AIHA. @AIHA. https://www.aiha.org/about-aiha
2. AIHA. (2023) Establishing a Process for the Setting of Real-Time Detection System Alarms. Falls Church, VA,
AIHA.org. https://aiha-assets.sfo2.digitaloceanspaces.com/AIHA/resources/White-Papers/Establishing-a-Pro-
cess-for-the-Setting-of-Real-Time-Detection-System-Alarms-White-Paper.pdf
3. ACGIH. (2023) 2023 Guide to Occupational Exposure Values. American Conference of Governmental Indus-
trial Hygienists.
4. ACGIH. (2023) 2023 TLVs and BEIs. American Conference of Governmental Industrial Hygienists.
5. AIHA. (2015a). Body of Knowledge: Field Use of Direct-Reading Instruments for Detection of Gases and Va-
pors.
6. American Industrial Hygiene Association.
7. AIHA. (2020a). Competency Framework: Field Use of Direct-Reading Instruments for…. @AIHA.
https://www.aiha.
8. org/education/frameworks/competency-framework-field-use-direct-reading-instruments
9. AIHA. (2021b). Reporting-Specifications-for-Real-Time-Detection-Fact-Sheet.pdf. Version 2, August 5. Falls

Church, VA, AIHA. pages. https://aiha-assets.sfo2.digitaloceanspaces.com/AIHA/resources/Reporting-Specifi-
cations-for-
10. Real-Time-Detection-Fact-Sheet.pdf
11. AIHA. (2020b). Technical Framework: Guidance on Use of Direct Reading Instruments. Falls Church, VA,
AIHA.org. pages. https://irp-cdn.multiscreensite.com/3bf7944a/files/uploaded/Technical_Framework_Guid-
ance_on_Use_of_
12. Direct_Reading_Instruments.pdf
13. NIOSH, (2007) NIOSH Pocket Guide to Chemical Hazards. National Institute Occupational Safety and Health.
https://www.cdc.gov/niosh/docs/2005-149/pdfs/2005-149.pdf
14. OSHA. (2019a). Permissible Exposure Limits- Annotated Table Z-1 Retrieved April 22 2022 from
https://www.osha.
15. gov/annotated-pels/table-z-1
16. OSHA. (2019b). Permissible Exposure Limits – Annotated Tables. https://www.osha.gov/annotated-pels/table-

z-1
80
Real-time detection of Nitrogen Dioxide in mining atmospheres: issues,

concerns, setting alarms, and interpreting data
R.E. Henderson, BS, MBA

GfG Instrumentation, Inc. Ann Arbor, Michigan, USA
ABSTRACT: NO2 is a pervasively common toxic contaminant in mining atmospheres. Updates to ACGIH
TLV and USA NIOSH occupational exposure limits, as well as underground mining exposure limits in many
countries, have made alarm setting strategies more difficult, and sometimes challenge the measurement capa-
bilities of the instrument to provide accurate results. This technical paper will discuss the capabilities and lim-
itations of the sensors used to measure NO2 including accuracy, resolution, effects of interfering gases, and
the effects of temperature and humidity on readings.
The paper will include discussion and interpretation of downloaded results logged during NO2 exposure
events from the underground Yucca Mountain Nuclear Waste Repository in Nevada, USA. The paper will
dis-cuss strategies for setting the instantaneous low and high (Peak) alarms, as well as setting STEL and TWA
alarms used in standard as well as extended work shifts.
Interfering gases, as well environmental conditions such as temperature and humidity can affect the perfor-
mance of the electrochemical sensors used to provide readings, and interpretation of downloaded data can be
difficult. The paper will identify common mistakes, and how to improve the quality and accuracy of monitor-
ing results.
1. INTRODUCTION
Need for a formalized process for establishing fitness for purpose and setting direct reading Real Time
Detection System (RTDS) alarms.
With the advancement of detection technologies and miniaturization of components, real-time detection of
the hazards occupying interest of the safety and hygiene practitioner have expanded. Those same technologi-
cal advancements now allow linking sensing and detection system data to automatically mitigate the condi-
tions (ventilation controls) or alert people to immediate action (fire alarm strobes, audible alarms, and per-
sonal safety instruments). Initial action is most frequently taken on the basis of instantaneous readings rather
than on the basis of statutory time history exposure limit calculations (e.g., STEL and TWA). Real-time in-
struments generally include STEL and TWA alarm calculations, but in most cases, the initial alarm is trig-
gered by exceeding a real-time concentration limit. Unfortunately, there remains a lack of guidance for estab-
lishing a “fit for purpose” process for establishing and setting real-time alarms when the Occupational
Exposure Limit (OEL) does not provide direct guidance.
The standard of care establishes five steps:

Step 1 – Statement of Purpose in target of condition to detect.
Step 2 – Selection of Objectives (parameters) around the condition to detect.
Step 3 – Selection of the device and determination that it is fit for purpose.
Step 4 – Establishment of an Alarm Set Point and setting expectations in response.
81
Step 5 – Communication of the results and determination.
The White Paper is intended to frame a common process that reasonably covers configuring appropriate
alarms on various types of real-time detections systems. In doing so, the process should sequentially assist
with selecting the sensor and configuring the real-time system, including, but not limited to, its datalogging
function and alarm setpoint(s). The entire process, when correctly applied, will ensure the user correctly inter-
prets the condition(s) induced when an alarm is triggered and takes an associated action.
PRACTICAL EXAMPLES OF ALARM SETTING GUIDANCE IN ACTION
Setting alarms for direct-reading instruments (DRI) can be the most complex part of implementing a real-
time detection system (RTDS). Recent developments in exposure limit changes, emphasis on systematic re-
view of the capabilities of DRIs, advancements in sensor technology, and reduction in instrumentation cost
have prompted discussion in the occupational and environmental health and safety professional community
regarding how alarms set points or thresholds are determined. Factors such as increased adoption of statistical
analysis methods and changes in the social relationship between workers and employers have also contributed
to discussion of alarm setting process transparency.
New guidance available from AIHA (AHIA, 2023) provides practical guidance for setting alarms. This
process includes five key steps to determine the appropriate alarm settings based on a systematic analysis of
the signalling context. Each step is intended to functionally align multiple components, including RTDS fit-
ness for purpose, systematic review of the DRI, and stakeholder expectations to result in a system configura-
tion that meets the needs the party designing and deploying the RTDS in a specific exposure context.
Key to this process is the fundamental principle that each alarm must trigger a defined response – from a
human, mechanical, or information system. The purpose of an alarm is to prompt action as defined by the sce-
nario: any alarm that is not intended to result in an action only contributes to confusing recipients’ response to
alarm conditions.
2. CASE STUDY: USE OF REAL TIME PERSONAL MONITORS WITH DIRECT READING NO2
SENSORS IN UNDERGROUND WELDING OPERATIONS
The USA Department of Energy (DOE) operates a secure nuclear waste repository in the state of Nevada.
The site is a large-scale underground site excavated into a salt dome. The geology gives rise to the potential
presence of naturally occurring sulfides including hydrogen sulfide (H2S) and sulfur dioxide (SO2). Carbon
monoxide (CO) and nitrogen oxides (NOx) are potential contaminants caused by of heavy equipment exhaust
emissions. Other hazards include the potential presence of combustible gas and oxygen deficiency. Portable
real-time direct reading instruments (DRIs) are routinely used throughout the site to measure these hazards
and ensure that toxic exposure limits are not exceeded. Inert gas welding on stainless steel is a common pro-
cedure at the site.
Problem: The DRIs in use at the site are subject unstable readings, and frequent, suspected false alarms
due to suspected cross interference of contaminants on the sensors, and the alarm settings currently in use at
the site. Several different brands of multi and single gas direct reading instruments are used at the site to
measure NO2. The instrument alarms and datalogging parameters are not set in a uniform way, and instru-
ments are not operated according to a standardized procedure.
82
Step 1: Establish statement of purpose.
Inert gas welding on stainless steel is known to produce oxides of nitrogen (NOx). Nitrogen dioxide
(NO2) is the most persistent and has the lowest toxic exposure limit. DRIs used by workers in the welding
area are needed to document that the 8-hour time weighted average (TWA) and 15-minute short term expo-
sure limit (STEL) toxic exposure limits for any contaminants that may be potentially present are not ex-
ceeded. Peak alarms based on instantaneous readings are used to alert workers to stop operations and leave
the affected area. Datalogging functions are required to provide documentation that exposure limits are not
exceeded, as well as trending analysis of such alarm conditions to determine supplemental controls (such as
behavior modifications of the work, or supplemental engineering considerations).
Step 2: Selection of objectives.
The primary exposure concern is with nitrogen dioxide (NO2) due to a very low occupational exposure
limit (OEL) utilized by the company. NO2 is a Designation B (Health Monitoring) hazard based on this OEL
assignment. The objective of monitoring is warning the worker to move away when nitrogen dioxide expo-
sures exceed either the real-time peak, TWA, or STEL exposure limit.
Step 3: Selection of devices and determination of fitness for purpose.
The conditions at the site, context of the work environment, and the toxicological basis for the Occupa-
tional Exposure Limit (OEL) are considered in this step. NO2 sensors are installed in the mine’s existing four-
gas, five-gas, and single gas monitors. The available NO2 sensors are limited to those which are certified for
use in the DRIs. The NO2 sensors offer industry standard performance with regards to range and resolution of
0.1 ppm resolution over a measurement range of 0 – 50 ppm. The mine was unable to identify products with
significantly better performance than their currently implemented devices, which are made by a well-known
manufacturer.
A significant issue is the response of the NO2 sensor to interfering contaminants such as SO2 (which can
cause negative readings), as well as the effects of humidity on sensor performance. The AIHA Standardized
Equipment Specification Sheet (SESS) was used to help clarify these sensor performance issues. It was found
that the NO2 sensor self-adjusts to changes in ambient humidity, that CO and H2S do not affect the NO2 sen-
sor. While SO2 is potentially present at the site, it is not generally present at the same time as the NO2 pro-
duced by welding operations.
The facility operates continuously. The work-shift is typically 12 hours. TWA alarms are calculated on
either the projected, or the most recent elapsed 8-hour period. Based on historical exposure data, most emis-
sions are below the 8-hour TWA and STEL exposure limit values, with rare instances of exceedance.
The DOE facility in this example currently uses the 1994 American Conference of Governmental Hygien-
ists (ACGIH) Threshold Limit Value (TLV) for NO2 (1 ppm TWA and 3 ppm STEL). The facility seeks to
follow the most current ACGIH OEL of 0.2 ppm TWA. ACGIH has not established an NO2 STEL.
Capabilities and limitations of real-time electrochemical (EC) sensors that measure NO2.
83
Substance-specific electrochemical (EC) sensors detect gas by means of an amperometric electrochemical

reaction that produces a current output (signal) that is proportional to concentration. Gas diffusing into the
sensor reacts at the surface of the sensing electrode. The sensing electrode is made to catalyze a specific reac-
tion. Use of selective external filters, and modifications to the electrolyte are used to further limit cross sensi-
tivity. The reaction used to detect hydrogen sulfide provides a good example of how EC sensors measure gas.
Hydrogen sulfide is oxidized at the sensing electrode:

H2S + 4H2O → H2SO4 + 8H+ + 8e-
The counter electrode acts to balance out the reaction at the sensing electrode by reducing oxygen present
in the air to water:
2O2 + 8H+ + 8e- → 4H2O
Thus, the overall reaction is:

H2S + 2O2 → H2SO4
The signal output for a typical EC H2S sensor, in this case the City Technology 4HS H2S sensor, is 0.7 mA
/ ppm H2S.
For EC sensors used to measure reducing gases, the reducing gas oxidized at working electrode and the
overall reaction consumes oxygen. Typical reducing gases measured with EC sensors include H2S, CO, SO2,
PH3 and others. For sensors used to measure oxidizing gases, the oxidizing gas reduced at working electrode
and overall reaction generates oxygen. Typical oxidizing gases measured with EC sensors include Cl2, NO2,
O3, HCl. HF and ClO2.
Nitrogen dioxide detection reaction:

At the sensing electrode:
NO2 + 2H+ + 2e- → NO + H2O
At the counter electrode:

½ O2 + 2H+ + 2e- → H2O
Overall:
NO2 → NO + ½ O2
Oxidizing gas sensors are particularly affected by changes in humidity. Sudden changes in humidity can
cause "transients" in readings. The sensor generally stabilizes rapidly in the new humidity, but the stabilization
process can take several seconds to 30 seconds or more. Typically, the signal rises and falls an equal amount
above and below the sensor’s baseline signal during the humidity response. Thus, the effects of humidity fluc-
tuation on the sensor signal tend to cancel out over time. It is important to avoid breathing into the sensor or
touching with a sweaty hand.
Be aware of potential cross sensitivity issues!
Although EC sensors are designed to limit cross sensitivity, most sensors show a response to at least a few
interfering gases. Since the detection reaction of oxidizing and reducing gas sensors has the opposite polarity,
unless removed by filters or eliminated by some other means, exposure to oxidizing gas can cause a reducing
84
gas sensor to show a negative reading, while exposure to reducing gas can cause an oxidizing gas sensor to
similarly show a negative reading. Two notable examples are the effects of SO2 (a reducing gas) on NO2 sensors,
and the effects of NO2 (an oxidizing gas) on SO2 sensors. Incompatibility issues may make calibration difficult
or impossible. If an instrument equipped with both sensors is fresh air adjusted in the presence of one or the
other gas, care must be taken not to adjust the opposite sensor. It is often better to install incompatible sensors
in different instruments.
Signal to noise ratio of EC sensors
Electrochemical amperometric gas sensors have a background current in addition to the current from the
oxidation or reduction of the detected gas. This background current is commonly referred to as zero current or
“noise.” Noise is random (stochastic) fluctuation of the electrical signal around a central value. Noise is meas-
ured by the Root-Mean-Square (RMS) value of the fluctuations over time. The signal to noise ratio (SNR) is
defined as the average over time of the peak signal divided by the RMS noise. When significant, noise can
interfere with measurements at low gas concentrations.
Different sensors can have different performance characteristics, even when they are designed to measure
the same gas. Sources for zero current “noise” include anodizing or cathodic reduction of the working electrode,
electrochemical oxidation or reduction of the sensor electrolyte, electrolyte contamination by the working elec-
trode, and reduction of oxygen in the ambient air.
Most amperometric toxic gas sensors are based on a three-electrode design. The detection reaction occurs
at the sensing (working) electrode. The balancing reaction that completes the electric circuit occurs at the coun-
ter electrode. The reference electrode has a known, constant potential against which other potentials may be
measured. An additional auxiliary electrode is sometimes used to correct (or reduce) noise. The auxiliary elec-
trode has the same catalyst structure as the working electrode but is not in contact with the measuring gas. By
subtracting the current of the auxiliary electrode from the total current of the working electrode, noise is re-
moved from the measurement signal.
Usability of sensor at low concentrations defined by RMS noise. Some NO2 sensors are limited to 0.1 ppm
resolution. Other NO2 sensors offer better resolution down to 0.05ppm or 0.02 ppm. Having a more limited
measurement range is often the tradeoff for better resolution.
Historically, instantaneous “LOW” alarms for toxic contaminants were often set at the TWA exposure
limit (TWAEL). This is possible as long as the sensor limit of detection is above the adjusted OEL. It is not
possible if the exposure limit is less than the lower detection limit of the sensor, or if there are not sufficient
incremental steps between the Limit of Detection (LOD) and the alarm setting to avoid false alarms. As a re-
sult of technological limitations of the instruments at the site, the 1.0 ppm TLV-TWAEL for NO2 is too low
to be used as the alarm setting for the instantaneous LOW alarm.
Step 4: Establishing alarm setpoint and response process.
Through collaborative discussions, site management arrived at alarm value policies using the ACGIH
guidance that worker exposure levels may exceed 3 times the value of the TLV-TWA level for no more than
15-minutes at a time, on no more than 4 occasions spaced 1 hour apart during a workday, and under no cir-
cumstances should they exceed 5 times the value of the TLV-TWA level when measured as a 15-min TWA.
85
In addition, the 8-hour TWA is not to be exceeded for an 8-hour work period. Accordingly, the following
alarm settings for NO2 were specified for use at the site:
Low alarm: 3.0 ppm

High alarm: 5.0 ppm
STEL alarm: 3.0 ppm
TWA alarm: 1.0 ppm
Step 5: Summarizing the analysis for effective stakeholder communication.
A summary report was prepared and delivered to the client mining company, describing the results of the
analysis for fitness for purpose, what the instrument could accomplish across the four agents of concern,
(NO2, SO2, CO and H2S) and how the alarm set points and specific defined employee alarm responses re-
quired by the workers were determined.
The report was reviewed and distributed to the business line and division management as well as the facil-
ity superintendent. It formed the basis for a briefing of the OEHS staff, subcontractors, and operational staff.
This report provided key transparency for workers seeking information on their protection during dissemina-
tion of instructions on how to respond to alarms.
The analysis summary was also pre-emptively provided to the site regulator as evidence of ethical and
comprehensive examination of the topic given specific regulatory, technology, and field conditions that set
the context.
3. INTERPRETING DOWNLOADED MONITORING RESULTS
Define the objectives behind the use of your real time instruments and make sure DRI capabilities match
the requirements. Make sure that the instrument settings match the objectives. Make sure the readings provide
the required resolution and accuracy. Make sure to understand the potential effects of interfering analytes and
ambient conditions on readings, and make sure the datalogging record supports the objectives.
Study the DRI manual and other documentation! Practice downloading results and make sure you under-
stand how the instrument records information and calculates alarms. Make sure DRI users are trained and that
the instruments are used in accordance with program requirements. Fresh air adjust the DRI before each day’s
use. Test and calibrate the DRI according to statutory and manufacturer requirements. Ensure that settings are
correct and that users follow proper procedures. Document everything!
86
REFERENCES:
1. AIHA. (2021a). About AIHA. @AIHA. https://www.aiha.org/about-aiha
2. AIHA. (2023) Establishing a Process for the Setting of Real-Time Detection System Alarms. Falls Church, VA,
AIHA.org. https://aiha-assets.sfo2.digitaloceanspaces.com/AIHA/resources/White-Papers/Establishing-a-Pro-
cess-for-the-Setting-of-Real-Time-Detection-System-Alarms-White-Paper.pdf
3. ACGIH. (2023) 2023 Guide to Occupational Exposure Values. American Conference of Governmental Industrial
Hygienists.
4. ACGIH. (2023) 2023 TLVs and BEIs. American Conference of Governmental Industrial Hygienists.
5. AIHA. (2015a). Body of Knowledge: Field Use of Direct-Reading Instruments for Detection of Gases and Va-
pors.
6. American Industrial Hygiene Association.
7. AIHA. (2020a). Competency Framework: Field Use of Direct-Reading Instruments for…. @AIHA.
https://www.aiha.
8. org/education/frameworks/competency-framework-field-use-direct-reading-instruments
9. AIHA. (2021b). Reporting-Specifications-for-Real-Time-Detection-Fact-Sheet.pdf. Version 2, August 5. Falls

Church, VA, AIHA. pages. https://aiha-assets.sfo2.digitaloceanspaces.com/AIHA/resources/Reporting-Specifi-
cations-for-
10. Real-Time-Detection-Fact-Sheet.pdf
11. AIHA. (2020b). Technical Framework: Guidance on Use of Direct Reading Instruments. Falls Church, VA,
AIHA.org. pages. https://irp-cdn.multiscreensite.com/3bf7944a/files/uploaded/Technical_Framework_Guid-
ance_on_Use_of_
12. Direct_Reading_Instruments.pdf
13. NIOSH, (2007) NIOSH Pocket Guide to Chemical Hazards. National Institute Occupational Safety and Health.
https://www.cdc.gov/niosh/docs/2005-149/pdfs/2005-149.pdf
14. OSHA. (2019a). Permissible Exposure Limits- Annotated Table Z-1 Retrieved April 22 2022 from
https://www.osha.
15. gov/annotated-pels/table-z-1
16. OSHA. (2019b). Permissible Exposure Limits – Annotated Tables. https://www.osha.gov/annotated-pels/table-

z-1
87
Remote Safety Audits – Drawbacks and Hidden Treasures
K. van Zyl, H. Botha, M. Biffi

BBE Consulting
Abstract
Objective/Aims of the paper/project:
One of the many challenges posed by the pandemic was the prospect of placing a hard stop to safety audits.
In the grander scheme of things, this may not appear as severe as protecting workers’ health or halting pro-
duction, but, in the longer term, the lack of a degree of assurance would be a challenge to the Health and
Safety Boards of Mining Companies. Considering the potentially catastrophic impact an unwanted event
such as a fire and/or explosion can have, this is particularly relevant. It becomes clear that there is a definite
need for assurance that the systems, equipment and actions put in place to prevent and mitigate the outcomes
of these types of unwanted events are required.
Undertaking safety auditing activities under full or partial lockdown has proven challenging. Substituting the
assurance afforded by an on-site visit where the ‘tires can be kicked’ and a ‘feel’ for the site can be devel-
oped, checking boxes and verifying ‘compliance’ to standards and procedures do not provide the same sense
of comfort. In addition, the value of face-to-face interaction and ‘water cooler talk’ to gauge the ‘mood’ and
‘attitude’ towards the impact requirements have many brought about by the pandemic itself is challenging to
be gauged and understood.
Design/Methodology:
A positive aspect is the acceleration of connectivity into daily operational activities. This, combined with de-
veloping analytical tactics that evaluate the appropriateness, consistency, implementation, verification and
monitoring of the control systems, equipment and actions needed to prevent and mitigate the impact of cata-
strophic unwanted events; compensate to some extent absence form site. Analytical tactics include adequate
and structured hazard identification, assessment of their impact on operations, verification of implementation
of and adherence to procedures, protocols and instruments such as critical controls, management of change,
implementation of learnings from related incidents, etc.
Results/Outcomes:
This paper provides an overview of the activities involved in remote safety auditing, it describes some of the
lessons learned, and provides an insight into opportunities to improve auditing effectiveness once “normality”
is restored.
88
Assessment of innovative mitigating solutions for fires occurring on

conveyor belts in coal mines
W. Marx , A. Thomson , M. van der Bank, N. van der Merwe

Bluhm Burton Engineering (Pty) Ltd
ABSTRACT: Despite tremendous strides in technology to improve occupational safety, the risk of fire in an
underground mine still remains high. Even though mechanization has revolutionized the mining industry, it has
also introduced several new fire ignition sources such as friction. Aside from being an ever-present source of
fire in a coal mine, conveyor belts also give cause for deep concern about the devastation left in the wake of an
underground fire.
If, the mining industry was not yet fully cognizant of this risk, the directive issued by the DMRE was certain-
ly the wake-up call needed to give urgent attention to an issue that had not been prioritized for decades.
It was this directive that created the necessary drive within the mining industry to fast track the development
of innovative solutions to the risk posed by fire sources, including conveyor belts.
In a joint effort, BBE and Thungela Holdings Ltd sought to better understand, through simulations, the effect
of certain mitigation controls during a conveyor fire. Fire simulations were conducted bearing in mind that, in
most cases, there would be other critical controls preventing the ignition or the propagation of a fire under
normal circumstances. The effectiveness of the mitigations was assessed, based on how well the mitigation
control performed under the assumption that no other controls were installed.
The aim of this study was to determine the size of the affected area as well as the rate of fume spread caused
by a fire. The intent was not to determine the actual concentrations of gas that might be encountered during a
fire.
Simulations were conducted for several fire scenarios on different conveyor belt locations in three different
coal mines.
1. FIRE MODELLING METHODOLOGY
1.1 General simulation principle
Fire simulation software packages require the user to calculate values such as estimated burn times and con-
taminant release ratios, as well as to make assumptions regarding the stages of the fire. These calculations,
estimates, and assumptions are based on research and theory and may vary in real-life underground environ-
ments due to constant changes during a fire. Estimates of burn times for the simulations gravitate towards a
worst-case scenario (shorter burn times = higher heat release rates = higher gas concentrations). Therefore,
information gained from fire simulations is considered an indication of possible fire impacts rather than an exact
result but may be used for risk assessments.
89
Ventilation simulation software was utilised to ascertain the concentration of gases during the phases of a fire,
as well as to track the spread of gases and heat throughout the mine. The programme can predict any flow
reversals that may occur due to the heat released by the fire event.
1.2 Heat release rate (HRR)
The amount of heat produced by a fire depends on the heat of combustion, mass of material burnt and the
efficiency of combustion. The HRR (kW) is calculated by multiplying the heat of combustion (kJ/kg) by the
material burn rate (kg/s) for the fire event.
The HRR from an underground fire (Brake et al., 2015) governs most of the behaviour of a fire at any stage
therefore is critical to any fire modelling. In general, the stages of a fire are said to follow the type of curve
shown in Figure 1.1. However, there are numerous ways in which a fire starts and progresses, and these ulti-
mately determine the duration of a fire and its peak HRR.
Figure 1.1 Stages of a fire
The three principal stages of a fire are (Brake et al., 2013):

• Growth stage: the fire starts and intensifies, controlled by either the available fuel or oxygen. When a
fire is in a confined area, the end of the growth stage is normally characterised by flashover. Flashover
occurs when the combustible material is heated to a temperature at which it auto-ignites into flames.
• Peak intensity stage: the fire burns at maximum intensity and has the most impact on by-products of
combustion, air temperatures and ventilation systems.
• Decay stage: the fire loses intensity as it runs out of fuel.
The peak HRR for a typical conveyor fire is between 8 MW (Hansen et al., 2011) and 11 MW (Lowndes et al.,
2006) for a fire consuming 0.5 kg/s of the rubber conveyor belt (burn rate).
90
1.3 Oxygen rich fires
In the case of oxygen rich fires, there is more oxygen available in the atmosphere surrounding the fire than is
being consumed by the fire itself. Heat produced by the fire is transferred to unburned fuel surfaces through
convection and radiation, causing pyrolysis in a largely localized process near the flame. Much of the air
passing through the fire zone is not involved in combustion which results in moderate to high oxygen concen-
trations downstream from the fire.
An oxygen rich fire is defined by the concentration of oxygen in the ventilation stream downstream from the
fire. The software calculates a corresponding heat influx due to the fire by multiplying the number of cubic
feet per minute of oxygen lost through combustion, by the standard combustion ratio multiplier (437 BTU per
cubic feet of oxygen consumed). The software assumes a nominal ambient oxygen concentration of 21% in an
uncontaminated airflow. The software variable which is used to specify an oxygen rich fire is the minimum
oxygen concentration downstream from the fire, expressed as a percentage of oxygen.
Example calculation to establish (HRR) (considering oxygen percentage):
Normal oxygen concentration in air mixture is 21%
Assumed oxygen concentration leaving fire is 15%
Oxygen concentration consumed by fire is 6%.
The software calculates a corresponding heat influx due to the fire by multiplying the number of cubic feet per
minute of oxygen lost through combustion by the standard combustion ratio multiplier of 437 BTU per cubic
feet of oxygen consumed.
Oxygen consumed by the fire is 6% x 42377 ft³/minute (20 m³/s) of air = 2543 ft³/minute (1.2 m³/s) of oxygen.
The HRR from the fire will be 2543 ft³/minute of oxygen consumed x 437 BTU of heat = 1111141 BTU/minute
or 19534 kW of heat.
A fire of 19.5 MW is considered to be very large and means that it consumes approximately 100 to 120 meters
of conveyor belt in one hour.
1.4 Fuel rich fires
In the case of fuel rich fires, the fire zone is so large and so hot that the entire quantity of air passing through
it is heated to a temperature sufficient to cause pyrolysis of fuel. The heated air passing though the fire zone
accelerates pyrolysis of the fuel with which it comes into contact. The pyrolysed fuel burns intensely until all
available oxygen in the air stream is depleted. Pyrolysis of fuel continues but does not burn due to lack of
oxygen. According to the software manual, fuel rich fires are considered to be rare events. It is estimated that
less than 1% of mine fires reach the fuel rich state.
1.5 Contaminant flow rate
The contaminant flow is the rate of contaminant gases added to the system by the fire. This rate is calculated
based on the chemical composition of the combustible materials being modelled. Once the combustion chem-
istry of the material has been determined, the contaminant mass flow rates may be calculated.
91
The most prominent gases produced during a fire event are CO and CO2. CO is generally the more toxic
contaminant of the two. However, due to the large quantities produced, the CO2 contaminant level is also
considered in the analysis. No other gases are tracked.
The ratio at which CO and CO2 are produced during a fire is dependent on the fuel type and the availability of
oxygen and fuel. For these simulations, as a major part of the product that burns is rubber and fuel, a ratio of
7% CO and 93% CO2 is used.
2. FIRE MODELLING INPUTS AND RESULTS
2.1 Modelling scenarios
A decision was made with the Head of the Ventilation Department firstly to run fire simulations for critical
conveyor belts which are worst-case scenarios for affecting downstream areas. In addition, an LHD fire was
simulated at random locations along the tractor roads. These simulations initially assumed no mitigating con-
trols to determine the impact and affected areas. Subsequent simulations tested different types of mitigation
controls on the same scenarios. The mitigating controls in each case depended on the location and type of fire
simulated. The amount of fuel selected for each situation represented the highest and most realistic heat release
rates to indicate what can be expected in a worst-case scenario.
2.1.1 First fire scenario – A fire on a section conveyor belt drive or a secondary conveyor belt drive
This scenario assumed a fire on a main trunk conveyor belt drive forming part of the main intake to a single
section or large portion of the mine. It was further assumed that, over a 50 m distance, the top and bottom of
the belt was fully involved, i.e. 100 m of 1 500 mm wide conveyor belt with a total weight of 4 500 kg, along
with 2 000 kg of the surrounding coal, and a burn time of three hours. These simulations initially excluded
mitigating controls, followed by simulations of the same scenario with one or a combination of mitigating
controls implemented.
2.1.2 Second fire scenario – A fire on a main trunk conveyor belt drive
The second scenario was similar to the first one but it assumed that, over a 50 m distance, the top and bottom
of the belt was fully involved, i.e. 100 m of 1 800 mm wide conveyor belt with a total weight of 5 800 kg, along
with 2 000 kg of the surrounding coal, and a burn time of three hours. These simulations initially excluded
mitigating controls, followed by simulations of the same scenario with one or a combination of mitigating
2.1.3 Third fire scenario – A fire on a main trunk conveyor belt tail end
This scenario was similar to the second scenario with the fire located on the tail-end instead of the drive but
with the same fuel involved. This simulation initially excluded mitigating controls, followed by a simulation
of the same scenario with one of the mitigating controls implemented.
92
2.1.4 Fourth fire scenario – A fire along a main trunk conveyor belt
In this scenario, a fire was simulated at a random position along a main trunk conveyor belt. Again, it was
assumed that the composition of the fuel involved was 5 800 kg of rubber and 2 000 kg of coal. This simulation
initially excluded mitigating controls, followed by a simulation of the same scenario with one of the mitigating
2.1.5 Fifth fire scenario – A fire on a LHD travelling on a tractor road
In this scenario, a fire was simulated involving an LHD at a random position along a tractor road. It was
assumed that the composition of the fuel involved was 1 000 kg of rubber, 250L of diesel fuel and 2 000 kg of
coal. This simulation initially excluded mitigating controls, followed by a simulation of the same scenario with
one of the mitigating controls implemented.
2.2 Mitigation strategies
Although many different mitigation strategies were considered, there were only a few that were able to control
the growth as well as the fume spread of a big fire while having a minimal effect on the normal operational
ventilation circuit. For a strategy to be effective, it needs to be easy to install and even easier to maintain.
Ironically, strategies that have the greatest positive impact on a fire and its fume spread, hinder maintenance
and breakdown repairs. All the assessed strategies are discussed below with their impacts on a fire and the
overall ventilation circuit.
2.2.1 Mitigation strategy 1: Enclosure of conveyor belt drives
Figure 2-1 shows the layout developed for the enclosure of the belt drives. It entails closing the entire belt drive
area with walls and conveyor seals to contain the effects of a fire long enough for the work force in the affected
area to withdraw to the nearest refuge bay. The exact layout depends greatly on the type and position of the
belt drive, as well as the accessibility around it. In this case, it is critical to assess the position of all ventilation
controls for each belt drive installation, considering access for maintenance, inspections, breakdown repairs,
and firefighting.
For practicality, use was made of a regulator of a predetermined size for each enclosure. This regulator was
simulated to be permanently open. The volume of ventilation to be drawn through the regulator must be suffi-
cient to ensure a negative pressure is maintained on all the seals around the enclosure. A good basis for the
design is that the larger the enclosure: the more seals, the greater the leakage and the higher the required regu-
lator quantity will be.
However, it is not possible to implement this strategy on all conveyor drives, especially in areas with high intake
volumes. In main intake airways, it might cause a restriction which could result in differential pressure across
the enclosure. This, in turn, might affect the ventilation flow direction through the enclosed area in case of a
fire and possibly cause high pressure on the upstream side and low pressure on the downstream side of the
airlock, forcing contaminated air to be drawn through the air lock.
93
The path that fumes follow in this scenario may be controlled and directed to ensure that a fire is not missed by
an EMS sensor.
Installation of a manually operated spray system inside the enclosure which can be opened by a proto team in
case of a fire should be considered.
A secondary effect of the enclosure of the belt drive is that dust generated inside the enclosure is contained.
While this has a positive effect on the quality of ventilating air going to the sections, it may have adverse health
effects on belt patrollers inspecting the conveyor drive area. Dust control systems should be investigated for
these enclosures.
This strategy was seen as a “quick win” solution for speedy implementation without any significant changes
to the infrastructure.
Figure 2.2 Mitigation strategy 1 – Enclosure of the belt drive area
2.2.2 Mitigation Strategy 2: Segregation of the conveyor and tractor road from the intake
Figure 2 shows another layout used for the enclosure or segregation of the entire belt, associated drives, and
traveling road. It entails enclosing the entire conveyor belt and the tractor road from the panel breakaway to
the section cross brattices, by constructing two lines of return walls on either side of the enclosed area.
This particular layout was prepared for a scenario where two sections are mining within the same ventilation
district. With the air crossings over the section, it was considered to be the worst-case scenario in terms of
access, positioning, available differential pressure on return walls, and airway capacity.
Fire simulations showed that the walls on the left of the conveyor should be of lower resistance than the main
return walls on the right of the belt. The leakage through the return walls on the left assists in maintaining a
negative pressure towards the main return of the section. This negative pressure ensures that no fumes or dust
escape into the section intake beyond the point of segregation.
94
Alternatively, as shown in Figure 2.2, a second regulator may be installed inside this enclosure ventilating
directly into the main return. Although this ensures some degree of flow direction through the enclosed area,
which can be directed over an EMS sensor, it also requires a ventilation commitment which is not always
affordable.
Ventilation velocities in this area may be very low and, depending on where the fire starts, there may be a
significant delay in the time it takes the EMS system to detect signs of a fire, especially if the fire starts along
the conveyor belt. Early detection of a vehicle fire is more likely as the operator is typically in the vicinity.
As in strategy 1, dust will be contained in this area as well, posing a direct threat to employees carrying out belt
inspections or patrols. If this mitigation strategy is seen as viable, the scheduling of patrols and inspections
along belt roads should be restricted to maintenance shifts if possible.
This strategy is seen as a long-term solution and would have to be implemented as sections moved to new panels
or new ventilation districts.
Figure 2.2 Mitigation strategy 2 – Enclosure of the entire belt road for the length of the conveyor
2.2.3 Mitigation strategy 3: Fire door opening – short-circuiting ventilation to lower intake velocities
and increasing available evacuation time.
Figure 2 shows the mitigating controls developed for main intake airways when the fire position is at the con-
veyor drive, somewhere along the belt, or if a vehicle starts burning on the tractor road. This control relies
heavily on the available differential pressure on the return walls running along the belt or tractor road. For it to
be effective, the intake and the return air circuit must belong to the same ventilation district so that the short-
circuiting of ventilation causes a decrease in intake velocities.
95
This mitigation strategy does not reduce the effects of a fire but provides workers with additional evacuation
time due to the slowing of the fume spread from the fire although, in some cases, all the generated fumes are
drawn into the return airway.
For a coal mine where methane build-up, due to lower air flow as a result of short-circuiting, is possible, it is
recommended that the position and overall impact of this control be simulated for each specific area to ensure
that the short-circuiting of ventilation still allows for enough ventilation to the sections to prevent unwanted
flammable gas build-ups.
This mitigation can be implemented quite quickly. However, the method of activation of the fire door should
be taken into consideration.
Figure 2.3Mitigation strategy 3 – Short-circuiting ventilation into the return airway
2.2.4 Mitigation Strategy 3: Fire door opening – assessing the position.
Additional simulations for mitigation strategy 3 assessed the outcomes of changing the fire door position in
relation to the fire. Similar fire characteristics were applied in conjunction with either placing the opening fire
door upstream or downstream of the fire.
The simulations showed that the position of the door had no impact on the duration of fume spread. However,
it did have an influence on the temperatures and gas concentrations leaving the area. The gas concentrations
downstream of the fire should not change the anticipated evacuation strategy because the value of the infor-
mation obtained from this study lies in the accuracy of the fume travel patterns and speeds, rather than the gas
concentrations. The concentrations encountered depend on several variables such as available fuel, type of fuel,
available oxygen, reaction time and do not change the priority which is to move all employees to the nearest
refuge chamber in the shortest possible time.
96
Fire Door
Downstream
LHD Fire
Fire Door
Upstream
Figure 2.4 Zibulo colliery - Trunk NM4 LHD burning - Fire door position assessment based on concentrations at point A
Error! Not a valid bookmark self-reference. shows the difference in CO concentrations encountered at point
A showed in Figure 2.4.
CO Concentration (ppm) CO Concentration (ppm)

Fire Door Upstream - Readings at point A Fire Door Downstream - Readings at point A
700 450
416
614 400
600 603
366
350
500
431 300
400 408 250
300 200 188
150 146
216
200
100 81
100 73
50
1 13
0 0
Time (Min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 Time (Min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180
CO ppm 0 0 0 13 137 408 551 603 615 614 575 511 431 354 281 216 159 114 73 CO ppm 0 0 0 13 146 416 412 380 373 366 327 279 232 188 148 112 81 54 34
Figure 2.5 Trunk NM4 LHD burning with Mitigation strategy 3 implemented – Fire door position result comparison.
2.2.5 Mitigation Strategy 3: Fire door opening - with mitigation strategy 2 implemented.
This simulation was to determine the effect of opening a tractor door upstream, or downstream, of a fire in
conjunction with the strategy 2 layout.
97
Due to the two lines of return walls in this layout, the flow through the enclosed area is limited to the amount
of leakage on the high-quality walls only. Opening one fire door, either upstream or downstream of the fire,
had no effect on the overall simulation outcome as the flow through the area is limited to the leakage on the
line of walls without the fire door.
A different outcome was seen when a cross flow of ventilation was created by opening a fire door at the section
cross brattices on the right of the belt and, simultaneously, opening a fire door at the panel breakaway on the
left of the belt. This created a directional ventilation flow over the fire, moving the contaminants to the return
airway, while maintaining a negative pressure on the return walls on the right of the belt. Thus, leakage into the
section intake was prevented.
The layout in Figure 2.6 shows how the airflow is affected by opening the fire doors upstream and downstream
of the fire simultaneously. Due to the directional flow through the enclosed area, it is unlikely that a fire would
go unnoticed for an extended period if the EMS sensor were placed close to the fire door at the breakaway of
the panel.
Figure 2.6 Mitigation strategy 3 - Fire door upstream and downstream opening in strategy 2 layout - simulation outcome
2.2.6 Fire door activation
Activation of the fire door might be just as important as the position of the fire door itself. Advances in tech-
nology mean that there are now several ways in which an automated activation system can be integrated with
the mine Scada system. However, certain requirements do need to be met in order to select a specific system
for this purpose. These requirements are:
98
Accuracy
Any system which provides real-time monitoring information should be accurate enough to enable realistic
decisions, based on the data provided, to be made. The difference between an actual gas reading and the output
provided by an accurate sensor should be negligible. The output of the sensor may be expressed as a percentage
of full scale or as an absolute measurement. It is imperative to know the type of output generated by the sensor
before setting points for activation of a fire door.
Reliability
Reliability refers to the consistency of a measure meaning that the output from a sensor should not vary over
the course of its lifetime. External environmental conditions such as temperature and humidity should not in-
fluence its overall accuracy. Short- or long-term exposure to contaminants should be prevented.
Maintenance and calibration

Proper maintenance begins with keeping the monitor clean and free of obstructions to the sensors inside. Sen-
sors installed in or around conveyor drive areas require regular cleaning due to the amount of dust generated in
these areas. Depending on the design of the sensor, its accuracy may be greatly affected if it is not cleaned
regularly according to a maintenance schedule determined by OEM.
A sensor must be calibrated to OEM specifications. Due to the high levels of vibration and airborne pollutants
which could affect sensor accuracy, underground bump testing should be carried out more frequently in areas
such as conveyor drives.
“Fail safe” system

An EMS system may be unreliable and inaccurate if used in environments as dynamic as underground collieries
where service infrastructure needs to be extended on a weekly basis. For this reason, a “fail safe” system in
which a single sensor is covered by a backup or secondary sensor should be considered. Then a fire door would
be activated only when both sensors reached a pre-determined alarm set point as per a risk assessment. This
would assist greatly in minimizing false alarms.
Set points and response delays

It is recommended that the background levels for each installed sensor be determined by studying the trend for
the specific sensor over the course of at least 2 months. The “background noise” level should be set as the
sensor’s zero level and the alarm set points calculated from this baseline. Response delays should be set up so
that a vehicle operating close to the sensor does not trigger activation of the fire door due to a false alarm.
Interlocking of back up sensors

Interlocking of the sensors downstream with the primary sensor is recommended. In most cases, simulated
contaminants initially bypass the fire area but, once the fire door is opened, very few to no contaminants bypass
the area. This is controlled using proper response delays. However, there are areas where contaminants pollute
the intake air for the entire duration of the fire. The sensor or sensors in a downstream position from the primary
sensor are affected by the pollution making it possible for them to trigger the opening of a fire door at a down-
stream position. Opening the fire doors at other locations may have a different impact on the fume spread from
the actual fire position.
99
It should be noted that once the fire has been extinguished, the fire door will have to be closed manually, either
by pushing a button or by a proto team underground. Simulations show that the installed sensor is affected
differently when the door is closed from when it is opened. Sometimes a sensor, reaching a set activation or
trigger level, will be in fresh ventilation after the door is opened and, if the system is set up to close the fire
door automatically, it will do so at this point causing the whole process to start over.
The system should be set up to provide for the opening of other fire doors downstream from the fire, in case it
spreads beyond the position of the fire door which covers the conveyor drive.
Activation
Methods of activating fire doors include the following:
• Manual activation - an assessment of the ventilation and EMS system by trained personnel which indi-
cates the need to activate the door manually (push button)
• Automatic activation - determined by smoke sensor, CO sensor, thermal camera or linear heat detection
– a primary sensor, with at least one other secondary or back up sensor, reaches a pre-determined trigger
level after which a PLC activates the fire door.
• Expert monitoring – a holistic appraisal of a system to assess the effect of opening a fire door on the
other components of said system before activating the door. This can be managed by rudimentary AI or
adaptive learning technology.
2.2.7 Mitigation strategy 4: Bypassing the area where a fire occurs.
This mitigation strategy is feasible if a mine is divisible into separate ventilation districts due to the availability
of intake facilities. A downcast airway, planned in the optimal position, allows for the separation of ventilation
districts. This enables workers to bypass an area where a fire occurs and allows the entire workforce in a certain
area, or the whole mine, to evacuate from underground, or to minimize the size of the affected area.
Employees trapped underground in a refuge bay are at greater risk of being exposed to a variety of other dan-
gerous conditions. A full evacuation from underground is considered the optimum outcome.
The mine shown in Figure 2.7 has an opportunity to use a new downcast shaft in the North as a secondary
escape. Figure shows how smoke reaches point A after approximately 45 minutes. Given enough warning,
employees working on the East of the mine will be able to evacuate in fresh air through a second main devel-
opment panel and escape via the new Northern downcast shaft. Simulations showed that, due to the position of
the up and downcast facilities in the North, a fire originating on the main belt train starting from shaft bottom
would not affect sections operating on the West.
The options for mitigation strategies for North Main Diagonal 1 trunk conveyor belt train are limited because
this area is in a main intake and has no return airway running along the panel. Therefore, most of the mitigation
strategies discussed would not be effective for this specific area. This scenario should be carefully analyzed if
the option to evacuate all employees safely to the surface in case of a fire is considered
100
Figure 2.7 NMD1 Conveyor Belt Drive without controls – Fume spread through the mine for a 3 hour burn period – point A
Point A
New Down
Cast Shaft
NMD1 Drive
is affected after 45 minutes.
3. SUMMARY
The fire simulation of conveyor belts directly in the intake clearly shows that a fire involving any of these units
is severe in nature. If a conveyor fire is not addressed, it produces more heat and gases and burns for a longer
period. In some cases, depending on the air velocity in a specific position, a fire may turn the airflow around
unexpectedly, against the normal airflow direction. A fire on an inclined belt may spread upwards quickly and
burn vigorously as more belt material becomes involved. The rate of spread for a fire depends heavily on the
air velocity in a specific area.
The introduction of certain easy-to-install mitigating controls proved very effective in preventing the spread of
fumes in all scenarios simulated. Simulations showed that, when enclosed, smoke and fumes can be directed
into the return airways quickly and independently of people or systems so that, in the event of a fire going
unnoticed for a period, fumes would not be able to enter the intake airways. The added benefit of the enclosure
of the drive areas is downstream dust control. Any dust generated by the drive and its components is directed
into the return airways. A system combining the enclosure of the drive with a spray system is worth consider-
ation. As fumes are contained in the drive, the proto teams attending to the fire would be able to activate a
manually operated, valve-controlled system.
However, there are some distinct disadvantages to these types of controls which may prove their undoing and
make them unsuitable. For example, the enclosure of a conveyor belt might prevent the execution of critical
maintenance and lead to failures of parts and ultimately fires. Furthermore, maintenance crews or belt patrol-
lers assigned to these areas would be exposed to high dust levels contained within the belt drive area. Fire doors
or access doors should be positioned so that they do not prevent access for firefighting and breakdown repairs.
101
Opening of a fire or tractor door to the return in a critical area to allow short-circuiting of ventilation, to delay
fume spread and to provide a longer evacuation period, proved effective in most areas. However, these controls
should be implemented mainly where an intake and return facility belong to the same ventilation district. They
will be more effective when used in conjunction with other controls but should be used only after accurate
simulations have eliminated any unforeseen changes in ventilation due to a fire.
The simulated mitigation strategy in which the entire belt road is enclosed, also proved effective in preventing
fumes from spreading to working areas. However, this mitigation may cause a delay in the detection of a fire
due to the low velocities in the area. Moreover, maintenance or belt patrolling schedules may be adversely
affected due to the likelihood of team members being over-exposed to dust when entering the area.
The simulation for bypassing a fire showed that it is possible to reduce the time during which employees are
trapped underground. The outcome of the indicative simulation provides a method of addressing the absence
of a second egress, if considered viable for future planning. Given adequate warning, employees can be evac-
uated to a predetermined point in another ventilation district from which they can be evacuated to surface.
Burn times of fires greatly depend on factors such as response time, accessibility, material involved, air quan-
tity, In order to understand what happens in the preliminary stages of a fire so that adequate emergency proce-
dures might be put in place, assumptions regarding burn times (including heat release rates) and burn rates were
made. These were found to favour higher burn rates rather than longer burn times. Therefore, the focus of the
study was to understand the propagation of fumes during the first 30 minutes of a fire to determine evacuation
strategies.
For any evacuation strategy to be effective, proper monitoring of underground environmental conditions should
be in place to provide ample warning in the event of a rapidly spreading fire. Early warning in this assessment,
would allow employees to evacuate unexposed to the fire. However, inadequate warning might result in em-
ployees having to navigate to refuge chambers in harmful fumes and low to zero visibility.
All simulations were carried out on conveyor belts where large quantities of ventilating air are available. Alt-
hough this dilutes gas concentrations and reduces leaving air temperatures, it also spreads pollutants to more
areas in the mine. If the fire simulations had been carried out in areas with lower air quantities, gas concentra-
tions and temperatures would have increased drastically but would have been concentrated in smaller areas.
Conveyor belts are manufactured from various rubber compounds and the by-products of combustion differ.
Other factors such as the fire position, ventilation quantity and combination of fuel types also produce varying
results.
The simulations showed that, even though temperatures leaving the fire were high, a rapid reduction in temper-
ature occurred as the distance from the fire increased. In some scenarios, temperatures dropped back to normal
within 200 m from the fire source. This is due mainly to heat transfer to the surrounding coal and dilution by
ventilating air. However, when fires burn for longer periods of time, heat transfer to the coal eventually ceases
and temperatures on the downstream side of any fire are higher.
Even though there is merit in having these types of mitigation controls in place, they should never be seen as
the only line of defence but rather as an additional measure that works in conjunction with other controls. The
controls that were simulated will have the greatest impact during the early stages of a fire when there is the
102
possibility of a delay in response to the alarm by a control room operator or in moving a proto team to the
incident. Controls that prevent fire in the first place, such as flameproof or fire-retardant belting, linear heat
detection, should take precedence over any other controls. Similarly, a well-maintained early warning system
together with a workforce that is prepared for emergencies through training, drills and simulations should take
priority. A system capable of instantaneous fire suppression and a well-equipped means of firefighting should
be in place as well.
REFERENCES
1. BBE Technical report - 22073-NFR-001-R0 Thungela - Thungela Operations - Fire Simulations
2. DMRE Instruction reference number CIOM-INSTR.OH-01-2022
3. Ventilation and Occupational Environment Engineering in Mines – Edited by Prof. J.J.L du Plessis, 3rd Edition
Revised and Expanded, 2014
4. The mandatory COP was drawn up in accordance with the Guideline with the DMR reference number DMR
16/3/2/4-B3 issued by the Chief Inspector of Mines
103
Emergency Evacuation Lessons Learnt

D van der Walt
Rescue One
L V Wyk and M du Toit
Two Rivers Platinum
D Byleveldt
Glencore Eastern Chrome Mines
ABSTRACT: The Mine Health and Safety Act, 1996, section 2.1 (a) requires an employer to ensure that
the mine is designed, constructed, and equipped to provide conditions for safe operation and a healthy work
environment equipped with a communication system and other equipment necessary to achieve those condi-
tions. The Mine Health and Safety Inspectorate, 2011, requires a mandatory code of practice (COP) for emer-
gency preparedness and response. The responsibility is placed on the employer to assess risk and to implement
the necessary interventions and controls. Emergency preparedness and response has in recent years developed
from a very rudimentary practice to a science, where precise planning is required, as is the deployment of state-
of-the-art systems and technology. In this paper, Rescue One Equipment and Industry (Two Rivers Platinum
and Glencore Eastern Chrome Mines), have joined forces to share best practices and lessons learnt on the de-
ployment of emergency preparedness and response.
Design/Methodology:
Emergency preparedness and response, to be effective, requires an integrated and holistic approach that in-
cludes people, processes and procedures and systems and technology.
In this paper, best practice will be linked to actual case studies implemented on the mines, that illustrate the
effectiveness of these strategies. As part of the emergency preparedness and response processes and procedures,
fire risk management, in particular, plays a crucial role in the successful deployment of emergency preparedness
and response. Another critical factor is ensuring processes and procedures are in place that will support a life
sustainable refuge bay, and effective emergency preparedness and response. A well planned and executed fire
risk management process, together with proper training on systems and technology, as well as well thought
through processes and procedures, will ensure that a mine is ready for any emergency.
As part of the system and technology leg of this model, systems such as Lifeline® are very basic and rudi-
mentary, but plays a crucial role in effective emergency response and have saved countless lives over the years.
More complex technologies such as gas detection and early warning systems are gaining huge traction and
yielding tremendous benefits.
As for the people focus, the importance of a committed leadership and an attitude towards safety will be
emphasised, as will the importance of staff that are well trained on the processes, procedures, systems, and
technology that is deployed at a mine.
Best practices and lessons learnt in each of these areas will be explored.
Results/Outcomes:
Glencore Eastern Chrome Mines have in recent years managed to avoid fire related incidents through a
diligent, well-planned process to address fire risk management. Two Rivers Platinum has managed to avert a
serious incident when a fire occurred at one of their conveyor belts. Their carefully considered risk management
process, with applied procedures, systems, technology and well trained and committed personnel, proved to
render great results. These two case studies will be explored, and it will be illustrated how these best practices
are contributing to successful emergency preparedness and response. These case studies will highlight key
success factors that any mine can benefit from.
104
BACKGROUND
During the early stages of a fire, smoke and gaseous products, including carbon monoxide (CO), are pro-
duced and released into the mine atmosphere. If these contaminants are not detected in the early stages of
combustion, they can result in severe hazards to personnel underground in the mine. Some underground mine
fires are discovered in their early stages by mine personnel who see or smell smoke, then make a quick decision
to either investigate and fight the fire or escape. Early detection of fires can save lives and prevent infrastructure
damage. The challenge with early detection is the positioning of detectors and sensors relative to the ventilation
flow of the mine. Pearton, Marx and Willemsen (2018) indicate that fire risk management in underground
mines can be improved by applying holistic fire engineering principles. Pomroy, Carigiet and De Rosa (Conti,
Chasko & Wiehagen, 2005) reviewed MSHA investigative reports for underground coal mine fires that oc-
curred during 1978–1992 and 1990–1999, respectively. They found that over 75% of these fires were first
detected by mine workers who saw smoke, smelled smoke, or saw the fire start. It is imperative during an
underground emergency that all persons, no matter their location, be able to get quick notification of the emer-
gency. To this effect, an integrated approach to emergency preparedness and response is recommended.
A PROPOSED MODEL
Integrated approach to emergency preparedness and response

Figure 1 contains a graphic representation of the proposed integrated emergency preparedness and response
model.
Figure 1. Emergency Preparedness and Response Model
It is proposed that all these factors should be carefully considered to ensure effective fire risk management
and emergency preparedness and response.
Processes and procedures

Safely operating a mine is a fundamental principle. All procedures and COPs are to be adhered to. Management
should put the necessary structures in place to ensure safe operations as per legal requirements and beyond.
The ultimate aim is to save lives.
Technology and systems

Technology and systems are playing an increasingly more important role in underground mines. The deploy-
ment of SCSRs has been an integral part of emergency escape and rescue procedures from areas affected by
irrespirable atmospheres. Biffi and Sehlabana (2018) quote research done at the University of Kentucky on
SCSR use, where it was clear that miners were not confident in the use thereof. The correct use and training
on systems and technology cannot be overemphasized.
A Refuge Bay is essential in the event of fire, as it provides a place of safety for mine workers to congregate
until proto teams can rescue them. Appropriate and visible signage to refuge bays is critical. Refuge Bay
control panels, lights, sirens and fans should at all times be in peak condition in the event of a fire. The latest
technology provides for digital displays that indicates life of battery and provides crucial information on how
long the battery can continue powering the fan.
Lifeline® is guidance system that consists of a steel rope with cones that directs the mine worker to safety
in an emergency. Lifeline® is operational in most coal mines, and recently gained traction in hard rock mines.
105
This is a very rudimentary system, but has been proven to be the most effective method to help mine workers
escape in zero visibility.
Monitoring systems are essential for fire emergencies. Implementing and operating monitoring systems
require dedicated resources specifically skilled for the purpose of effective telemetric monitoring for the un-
wanted event of fire and gas release. A comprehensive systems approach must be followed where some of the
following aspects are crucial for successful and effective detection:
• Placement and positioning of sensors; air velocities and air traveling distances to sensors;
• Types of sensors to be used;
• Integration of detection devices other than toxic gas detectors;
• Alarm settings and ratios;
• Control room attendant proficiency;
• Evacuation and response procedures; and
• Maintenance of such systems.
Early warning systems allow for proactive evacuation. Frontline®, one such early warning system, is a type
of stench gas system. This system differs from the typical stench gas system as it releases a very strong grape
or citrus fragrance in the intake, which is non-toxic. Testing has shown that grape or citrus fragrance can be
detected in smoke. Depending on the ventilation flow in the mine, Frontline® could give mine workers 20-30
minutes head start to escape to a place of safety. In mining there is a place for rudimentary systems such as
Lifeline®, as well as for more technologically advanced systems such as Frontline® and other detection sys-
tems. It is about harnessing the best of both worlds.
People
People involved in an emergency, are likely the most important factor that can lead to success or disaster.
Emergency preparedness and response training is one of the most important variables that can lead to safe
evacuation in case of a fire in an underground mine. Biffi and Sehlabana (2018) quotes research done by Brnich
et al (1999) where forty-six miners were interviewed after escape experiences. The miners involved in the
escape experiences admitted to panicking. Some miners admitted to not knowing what to do at the time. Can
miners afford not to know what to do? One life lost in an incident is one too many. Training can ensure that
mining personnel perform an evacuation without having to think about what they must do. Realistic, experien-
tial training, or in other words real-life evacuation simulations are crucial. Effective emergency preparedness
and response training can have a significant impact in removing a feeling of panic and uncertainty.
Management and leadership buy-in is a critical success factor. Haas et al (2014) indicates that an evident
gap illustrated by trainers was the need for mine high-ranking managers or safety personnel to have presence
or participation during mine safety training. This displays their commitment and sets the tone for safety in the
mine. They also recommend that state organisations, such as the Department of Mineral Resources and Energy
(DMRE) in the South African context, also be present during training. This would contribute to a better under-
standing of the state of the mining industry and practical challenges experienced. Their presence and under-
standing could contribute to better cooperation between the mines and the DMRE. It is important to note that
working safe is also a mindset, as each mine worker must take safety seriously and look out for him/herself, as
well as for the welfare of fellow mine workers. A safety mindset starts at the top. Leadership by example. A
major element of safety is taking proper care of emergency response equipment daily. Should a mine worker
walk past a defective emergency device, he/she should take the necessary steps to either report and or repair the
device. Each leader, manager and mine worker should have a mindset to look after any equipment designed to
save lives in event of an emergency such as a fire. This principle was recently accentuated at the Sacesha
Conference 2023 by Inus Labuschagne, with specific reference to Lifeline®.
Kowalski et al (2010), did research on three actual mine fires, considering leadership roles in these emer-
gency escape scenarios. They found that a profile of leadership in crisis emerged from the analysis of these
escapes. The data suggests several characteristics based on the behaviour of the leaders. The leader of each
escape may be described as an aware, knowledgeable person or as an individual who is alert to his environment,
attentive, and discerning. The leader notices things more so than other mining personnel. This astute quality
is probably not limited to the mine environment or to crisis circumstances. This individual is naturally curious
106
all the time and is the type of individual who excels at incidental learning. Each of the leaders retained infor-
mation that was instrumental to the escapes.
Traits that stood out were:
• They recalled specific details and repeatedly referred to the fact that they "knew" through information or
deduction;
• They took charge;
• They were decisive, yet flexible. They made decisions; yet if circumstances changed, they adapted;
• They were open to input from others;
• Effective leaders seemed to have a calming effect on their group; and
• There was a Logic to their leadership. Decisions were appropriate and congruent with the available
information.
CASE STUDIES
Two case studies will be used to illustrate key lessons learnt by industry. Two Rivers Platinum will reflect
on a fire incident they had and the factors that contributed to the mine getting the situation under control quickly
and effectively. Glencore Eastern Chrome Mines will reflect on what they have done to avert any serious
incidents to date.
Two Rivers Platinum (TRP)

Two Rivers Mine recently had an underground fire on a sacrificial conveyor which propagated for 15.8
meters (both top and bottom belt) from the point of ignition. Due to the swift action taken by the involved
parties, functional documented procedures, sound communication and technology in place, the fire was suc-
cessfully extinguished, and nobody was injured at the time of the incident. The model in Figure 1 addresses the
key aspects of holistic emergency evacuation. How TRP applied this model, will now be extrapolated upon.
An inclusive approach is followed at Two Rivers Mine when it comes to fire preparedness and response
management. Planning for and preparing for unforeseen events such as fires are crucial as part of any mine’s
fire risk management process. On the day of the fire incident, some control measures failed, but due to the
overall combined approach, the other control measures in place assisted with the speedy and well controlled
emergency response management.
Elements that form part of the Two Rivers Mine holistic approach to fire preparedness and response include:
§ Processes and procedures
Having well thought through processes and procedures are a great foundation for emergency preparedness
and response. These should be developed by consulting the right stakeholders that can add real value to the
content. TRP has the following in place:
• A well-established risk management system where fires were identified as one of the top ten risks
and sound management plans were developed focussing specifically on fire risk management;
• Code of Practices and Operating procedures covering the management of fire risks, technology in
fire detection and suppression, competencies, and verification inspection of fire control measures;
• Internal audits and inspections on fire management systems; and
• External audits and inspections on fire management systems.
Regular audits provide for quality feedback on processes that can enhance these processes.
§ Systems and technology
There are three key factors when it comes to systems and technology, namely that the persons responsible
to use these systems should be fully competent in doing so, correct installation of such systems and technology,
and the systems and technologies should be well maintained and kept current. TRP has appointed Original
107
Equipment Manufacturers (OEM’s) to install and maintain fire systems and technology comprising of Fire
Detection Systems (FDS), Fire Suppression Systems (FSS) and Fire Fighting Systems (FFS)).
Realtime data on gas detection is a major success factor. TRP has real-time gas detector sensor data readings,
(positioned at heads and tails of all conveyor belts) available on the SCADA system in the control rooms. A
minimum of two persons per work area underground must have Gas Detection Instruments (GDI’s) with them
at all times. 98% of conveyor belts are grade F as prescribed by DMRE. Grade F rubber belting is fire resistant
and has added properties that increase the ignition temperature and increase the resistance to fire. Conveyor
belts typically will start burning at temperatures more than 325 ºC. Furthermore, TRP has interlock systems on
conveyors, belt slip detection and belt aligning devices. These are also DMRE requirements. In an event of
fire, the belt stops immediately and contains the fire to the area where it occurred.
Automated fire suppression systems are installed at the Heads and Drives on conveyor belts. This allows
for quick suppression of fires without putting people in harm’s way. In addition, fire extinguishers and fire
hydrants are placed at pre-determined positions and regularly checked, maintained, and over inspected. One
channel on the radio system is allocated to emergencies. Having a dedicated emergency communication chan-
nel on the radio system, removes unnecessary chatter during an emergency.
§ People
TRP nurtures a workplace culture that embrace safety. Skeepers and Mgohwa (2015) highlight that the follow-
ing are essential to the development of a positive safety culture:
1. Communicating company values;
2. Demonstrating leadership;
3. Clarifying required and expected behaviour;
4. Personalising safety outcomes;
5. Developing positive safety attitudes; and,
6. Engaging and owning safety responsibilities and accountabilities.
Employees are trained in the emergency procedures together with scheduled emergency drills. All shifts com-
plete monthly underground drills, and surface drills are conducted twice a year. All Patterson D-Band manage-
ment personnel complete the Mines Rescue Services (MRS) Emergency Control Room Management training
which is renewed every three years. TRP has functional and well-trained proto team and three fire teams on
site, and a contract with MRS. All fire teams are trained by Emergency Training Solutions (ETS).
§ Lessons Learnt from a Fire Incident
When it comes to testing the adequacy of implemented control measures, often, the only way to determine how
well your system works, is through an actual emergency itself. It is therefore very important that lessons learnt
are captured, shared and embraced. With the TRP Mine fire incident, most of the control measures proved to
function very well during the emergency, but there were some elements that did not work as anticipated and
where lessons were learnt, and improvements made within the current fire preparedness and response manage-
ment system.
On 29 May 2022 at 22:10 it was reported to the Main Shaft control room that smoke was observed at one of
the 14 underground working levels. A sacrificial conveyor belt caught fire during blasting time and when the
blasting fumes cleared, the smoke from the conveyor belt continued to contaminate the section of the mine
where the fire detection CO and NO2 sensors were reporting high readings. Employees were sent underground
after the one-hour blasting re-entry period has lapsed. The high gas readings on the SCADA system were over-
looked and the audio alarm linked to the SCADA system was not functioning on the day. It was only on the
arrival of the employees to the section that the smoke from the burning conveyor was detected.
This is consistent with research by Pomroy et al (Conti, Chasko & Wiehagen, 2005) that found that over
75% of these fires were first detected by mine workers who saw smoke, smelled smoke, or saw the fire start.
The TRP crew promptly withdrew to the Refuge Bay and reported the incident immediately to the control room.
Ever since the incident, TRP has a dedicated internal contractor that maintain and monitors the functionality
of the Fire detection system. This is crucial to avoid any future malfunction. Upon receiving the emergency
108
call, the control room operator immediately followed the instructions as per the Standard Operating Procedure
for Emergencies and contacted the required persons. The VOHE Manager monitored the gas readings closely
and liaised with the Mining Leader. At the time of initial reporting, the CO reading was 188ppm and NO2 3ppm.
During this time an electrician was tasked to investigate the reason for the power failures in the reported area,
staying within fresh air intakes and in continuous communication with the control room. During his investiga-
tion it was identified that the dip belt had burned. On the conformation of the burned belt, the Emergency
Procedure was initiated, and the Emergency Control Room established.
After all relevant information was gathered and analysed, the manager in control decided on a strategy, the
team was briefed, instructions were communicated and logged, and there after the Proto Team was deployed to
the area. At this time, the beacon heads indicated a CO reading of 144 ppm, and NO2 reading of 2 ppm. Specific
instructions to the Proto team were to do a rescue and evacuation in all identified and affected Rescue Bays,
and not to enter into areas with irrespirable atmosphere (due to gas readings over the legal limits), to check
around the belt and to report all findings and to use the dedicated emergency channel for communication. The
guidelines of Mines Rescue Services stipulates that no team may be deployed without a backup team during an
incident. This includes conditions such as entering irrespirable atmospheres (donning of the breathing appa-
ratus), hot environment as guided in the Emergency Heat Stress Index, Rescues, Recoveries, Rope work, Re-
connoiters and if a risk assessment indicate risk to the team deployed. The second proto team (from Thorncliffe
mine) was called out to be the backup team for the Two Rivers team. Within an hour, on arrival, the Thorncliffe
team was briefed on information gathered so far and they were instructed to proceed underground to an indi-
cated refuge bay and establish a fresh air base with communications to surface and the operating team. Backup
communications to the operating team was the gene phones.
Only when the Fresh Air Base (FAB) was established and the backup team stationed in the FAB, did the Two
Rivers Proto Team go into Oxygen and proceeded to investigate the area of the belt fire. The gas measurements
taken by the two Rivers Proto team were CO - 100pm, CO2 - 2000ppm, Oxygen - 20.2% and a temperature of
29°C wet bulb / 30°C dry bulb and after roles were reversed the Thorncliffe Proto team was instructed to take
a reading on a designated point and the gas measurements reported were CO 90ppm, CO2 1900ppm, Oxygen
20.2% and temperature of 30/31 °C. Only at 6:40 that morning did the Proto Teams return to surface. Through-
out the incident, clear communication between the various parties took place and all parties were aware of what
their duties were within this emergency.
MRS Training was instrumental in establishing a dedicated control center, and having dedicated procedures
in place. Because management and staff were trained, everyone knew what to do. This is a key success factor.
In the Emergency Preparedness and Response COP, TRP has a clear and well-defined checklist of the roles and
responsibilities. This is done per function/discipline. Again, because of training on the COP, everyone knew
their roles and responsibility. As per MRS training, a single person was in control of the incident and were
continually advised by the VOHE Manager, MRS Manager and other key personnel that was involved as
deemed necessary.
The training provided by MRS to the Brigade members and to identified, relevant and qualified Managers,
on how to manage the incident from the control room during an emergency, proved invaluable and were cited
as a major success factors. It contributed to clear roles of responsibilities clear communication, and decisive
action.
On the response to the emergency no substantial gaps were identified. Communication was fluent and the emer-
gency was well managed. However, on Fire preparedness some gaps were identified and were addressed im-
mediately after the incident. Gaps identified included the following:
• Employees proceeded underground without confirming if the gas readings on the SCADA system were
cleared. The alarm also did not work on the day to warn the control room operators that high gas readings
still prevailed. As mentioned before, a dedicated contractor has been deployed to maintain and monitor
the system to avoid any future malfunction;
• Gas sensor heads were placed at all head- and tail positions on all belts, however, the positioning did
not cater for the sacrificial belt. Due to the ventilation arrangement, the high gas readings were picked
up by the gas sensors situated on the tail ends from strike belts within the ventilation “flow”;
• The strike conveyor was not stopped after end of shift. The belt appeared to be off to the belt attendants,
not knowing that the belt was actually “stuck” due to a spillage at the other side of the belt;
109
• The interlock system between the strike belt and the sacrificial belt (where the fire originated due to
product build-up) did not work. The slight change in the tipping arrangement was overlooked and a
“management of change” never performed. It was also found that the speed switch which should trip
the conveyor when slippage occurs, was connected and in order, but the Simocode which interfaces with
the starter was not programmed to detect the speed switch;
• Automated fire suppression systems were only positioned at the head- and drive positions of the belt
and not in remote areas such as the sacrificial belt, thereby leaving the sacrificial belt at risk in times
when no one is around (only manual firefighting systems were available at the sacrificial belt);
• Although most of the conveyor belt consist of type F belt, a portion that burned on the day of the fire
was not Type F and had poor flame resistance properties.
From the above it is clear that post an incident, these conditions must be reflected on and corrected for future
prevention. The gaps have consequently been addressed. This fire had the potential to have very serious conse-
quences. Disastrous consequences were averted because Two Rivers Mine applies a holistic fire prevention and
management approach. Should one aspect fail within the approach, other aspects thereof will be in place to
address the issue at hand. The other, and maybe the most critical aspect, is the people aspect. On the day, clear,
well organized, communication took place between all the relevant parties.
Glencore Eastern Chrome Mines
Glencore Eastern Chrome Mines (ECM) is situated in the Steelpoort Area. It consists of three mines. The
oldest is Thorncliffe Mine, that started in 1997. Magareng Mine started in 2008, and the youngest is Helena
Mine, which started in 2004. In this case study, the preventive measures that Eastern Chrome Mines apply in
dealing with emergency evacuation and preparedness will be discussed using the model described in Figure 1
above.
3.2.1 Processes and Procedures
ECM places a lot of value on risk management. Glencore Eastern Chrome Mines implemented the Code of
Practice for Emergency Preparedness at each operation. Each Code of Practice have supporting procedures,
documents and Risk Assessments linked to their specific conditions. As part of emergency preparedness, the
operation need to ensure that employees are safe, and should there be an emergency, the company can ensure
that personnel can be removed either from the working place to a place of safety or evacuate employees to
surface should it be safe to do so. Secondary escape routes have been established between all three mines. All
three mines are linked together. Evacuation can be done via any of the three mines. Escape routes are demar-
cated and inspected monthly. The Code of Practices are revised on a three yearly basis, or when required after
a serious incident or when significant changes to procedures, mining methods or layouts are introduced.
Glencore Eastern Chrome Mines have Catastrophic Hazard Management plans in place. This will be appli-
cable to catastrophic events that can lead to five or more fatalities in one single event such as fires and explo-
sions, Fall of Ground (FOG), Trackless Mobile Machinery (TMM) related incidents etc. Each mines have an
emergency control room which will be activated during any emergency and the key role-players will be present
as required by the emergency control center procedure depending on the type of emergency.
Glencore Eastern Chrome Mines have the following agreements in place to ensure fast and efficient emer-
gency response:
• The Eastern Chrome Mines Occupational Health clinic is on site and is equipped to handle any
emergency. A service level agreement with Netcare 911 is in place, with a full time Occupational Medical
Practitioner, trained and qualified Basic Life Support, Intermediate Life Support personnel, as well as an
Advanced life support ambulance which is available 24/7. An additional Advanced Life Support
Paramedic is available in case of emergencies;
• Part of Netcare 911 SLA is the availability of air lifting services should a patient be required to be airlifted
to one of the preferred Hospitals for specialized treatment;
• Eastern Chrome Mines have two competent and trained proto teams with the proto room situated next to
the on-site clinic at Thorncliffe mine and equipped with the required equipment as stipulated in Chapter
16.5 of the MHSA; and
110
• Eastern Chrome Mines operates under a contract with the Mines Rescue Services which are based in
Steelpoort as an A-class member mines in case of any emergencies.
3.2.2 Systems and Technology
The following has historically been in place:

• All three mines have equipped and life sustainable refuge chambers underground for each section;
• All employees required to use continuous gas detection instruments are trained and found competent
before issued with the instrument;
• All lamp room equipment is serviced and maintained as per Original Equipment Manufacturer (OEM)
specifications and each Lamp Room has its own appointed Lampsman to ensure compliance;
• Underground fire detection systems are installed and maintained by the OEM as per service level agree-
ment; and
• Underground radio communication is used to communicate to the control room and a dedicated emer-
gency channel is provided for emergency use only.
ECM evaluated the above and came to the following conclusions. The systems deployed in the mines were
not able to communicate directly to the employees working on the face. The mine only had radio communica-
tion up to the tips in each section, thus relying on the Tip Attendant to call or warm employees. Should the radio
communication fail during an emergency, no warning could be given. To address this gap, Audio-Visual (AV)
Alarms were installed (Figure 2).
Figure 2. Photo of Audio-Visual Alarm System
These provide high performance visible warnings, which is activated from the surface control room in case
of an emergency to enable employees to evacuate to a safe place. The LED strip lights are connected to the AV
Alarm which consist of three lights installed on dip. The light will flash red once activated from the surface
control room. The AV alarms were installed in the middle panel of each section and as close as possible to the
face. Persons working above or below the unit will hear or see the alarm and react accordingly. The OEM
maintains and moves the alarm system forward monthly.
Eastern Chrome Mines values the lives of their employees, and consequently decided that improved emer-
gency evacuation systems were needed to provide for effective emergency evacuation. Learnings from the coal
industry was adopted. In addition to the AV Alarm systems, ECM decided to install Lifeline®. Lifeline® is a
guide rope that is traditionally used in coal mines as a system to help miners escape to a place of safety in zero
visibility. Getting mining personnel to a place of safety in zero visibility is crucial, but equally important is
avoiding a situation where mining personnel need to evacuate in zero visibility. To counter this gap, Eastern
Chrome Mines decided to look at early warning systems. To this effect, Eastern Chrome Mines have installed
the Frontline® fragrant system. This fragrant system provides a unique grape like smell which can be smelled
in the air containing smoke. The system is activated from the surface control room in case of a fire emergency
to enable employees to evacuate to a safe place.
A test was run on these systems to measure its efficacy. The Frontline® units were placed at two positions
(as per the yellow dots on the mine map in figure 3). Proto Team members were placed at strategic positions
(as per the red dots on the map). The Frontline® gas canisters were dispersed at the same time. The distance
between canister 1 and 2 was 1800m, and the distance from Canister 2 to the last point of through vent was
2200m. The canisters were dispersed at18:57 and last person at the furthest point smelled the fragrance at 19:09.
In other words, within 12 minutes the fragrant was smelled throughout the shaft including ROM Belts, the
111
Workshop and Workings. This proved to be a very effective early warning in the event of fire, providing mining
personnel an opportunity to evacuate before zero visibility.
Figure 3. Frontline® Early Warning Fragrant System Test
Additional measures that were implemented included:

• 2nd Escape routes were established which link all three mines. This allows for evacuation from any one
of the mines if needed;
• Fiber cable was installed and a ring feed using the escape routes, linked to all refuge chambers; and
• Additional telephone and cameras were installed in each refuge chamber to always ensure communica-
tion during an emergency.
3.2.3 People
People using these systems are the most important factor to ensure success. Training is considered as a
fundamental activity. All shafts have trained level 1 first aiders on surface and underground to attend to injured
employees if not life threatening. Monthly emergency evacuation drills are conducted per section.
Annual mine wide evacuation drills are planned to assess the employee’s reaction when evacuation systems
are activated. With these drills, paramedics and Proto team response times are also tested to see if there is room
for improvement. Important to note is that all shortcomings and findings are actioned and addressed to ensure
employees are well equipped with the knowledge and skills in case of an emergency. On initial and periodic
refresher training all employees are trained on the AV alarms sound and light as well as the smell of the fragrant
system. Frontline® training is conducted in the Training Center whereby the facilitator has a bottle of the
fragrant available for all learners to smell during the emergency evacuation training. This allows the employees
to get familiar with the smell.
Lifeline® training is conducted in ECM’s underground training center whereby employees can see how it
should be installed and to use the guide rope to direct them towards the Refuge Chamber (refer to Figure 4).
This allows employees to feel the cone direction and to ensure that it is correctly installed.
112
Figure 4. Lifeline®
Self-contained self-rescuer (SCSR) Pac training is conducted for all underground employees. ECM makes
use of training units to allow each employee to demonstrate the donning process as part of the assessment. Their
training center is equipped with a breathing simulator to allow employees to get the feel for the rescue pac
breathing tube, how-to breath with the Rescue Pac and to feel the heat of the chemicals while you breath. During
the annual mine wide evacuation drill, ECM issues expired SCSR units to identified employees to physically
don during the exercise to make the drill as realistic as possible. Important to note is that mines should specify
in their training material when an employee is required to don his rescue pac.
Biffi and Sehlabana (2018) emphasise the importance of hands-on training in the complete donning of all
types of SCSRs in use in the mine, as well as hands on training in transferring between self-rescue devices at
change-over stations.
Biffi and Sehlabana (2018) further quote research done by Brnich et al (1999). Forty-six miners were inter-
viewed after escape experiences. Significant observations included:
• None of the miners had worn a SCSR prior to the emergency;
• Most miners indicated that they were not proficient in donning an SCSR;
• Some miners delayed donning their SCSR because they were aware of the limited duration of the unit;
• About 67% indicated they donned the SCSR in smoke; and,
• About 59% admitted to removing the mouthpiece or to breathing around it in smoke once the felt the
resistance to breathing had become excessive.
Leadership plays an integral part of our employees’ safety and wellbeing. Leaders need to equip their em-
ployees with the correct knowledge and skills to ensure that they understand procedures and they know what to
do in case of an emergency. One can install or implement any state-of-the-art emergency evacuation system but
if the employees are not educated on the system and what actions to take, the system will not be effective, and
the risk of losing lives in an emergency situation becomes significant. Kowalski-Trakofler et al (2010) conclude
that hands-on training in realistic situations is the best way for miners to become prepared and confident.
CONCLUSION
Pearton, Marx and Willemsen (2018) indicate that fire risk management in underground mines can be im-
proved by applying holistic fire engineering principles. Pomroy, Carigiet and De Rosa (Conti, Chasko &
Wiehagen, 2005) reviewed MSHA investigative reports for underground coal mine fires that occurred during
1978–1992 and 1990–1999, respectively. They found that over 75% of these fires were first detected by mine
workers who saw smoke, smelled smoke, or saw the fire start. It is imperative during an underground emer-
gency that all persons, no matter their location, be able to get quick notification of the emergency. To this effect,
a holistic approach to emergency preparedness and response is recommended as per figure 1.
The foundation is good processes, policies and procedures, backed up by systems and technology that is
correctly installed and functional. And most importantly, staff must be trained on using these systems. At the
2023 Sacesha Conference, Inus Labuschagne highlighted some fire incident lessons learnt, and said that Life-
line® is as the word says, a “life line”. And that all staff should be trained to respect the function of this device,
113
it will save lives. It is recommended that newer technology pertaining to the use of stench gas released close to
workplaces be implemented as early warning to facilitate evacuation from risk areas while visibility is not yet
impaired. Early detection systems can play a major role in early evacuation once a fire has commenced and
can help mine workers escape to a place of safety without having to don SCSRs or use Lifeline®.
Management buy-in is thus a critical success factor. Haas et al (2014) indicates that an evident gap illustrated
by trainers was the need for mine high-ranking managers or safety personnel to have presence or participation
during mine safety training. This displays their commitment and sets the tone for safety in the mine. Training
if staff and management was a major success factor raised in both case studies.
ACKNOWLEDGEMENTS
Rescue One Equipment would like to extend our heartfelt appreciation to Glencore Eastern Chrome Mines
and Two Rivers Platinum for their openness in sharing best practices on fire risk management and emergency
evacuation. And for the inputs from Mines Rescue Services and ETS. Both rendering an invaluable services to
the mining industry.
REFERENCES
Biffi, M. and Sehlabana. M. (2018). Self-Contained Self-Rescuer Training – Doing more with more. The Mine Ventilation Society
of South Africa, 2018 Conference.
Brnich, M.J. Jr; Vaught, C. Calhoun, R. (1999). “I can’t get enough air! Proper Self-Contained Self-Rescuer Usage – Instructor’s
Copy. US Department of Health and Human Services, Centres for Disease Control and Prevention, National Institute for Safety
and Health Bulletin, October 1999.
Biffi, M. and Sehlabana. M. (2018). Self-Contained Self-Rescuer Training – Doing more with more. The Mine Ventilation Society
of South Africa, 2018 Conference.
Brnich, M.J. Jr; Vaught, C. Calhoun, R. (1999). “I can’t get enough air! Proper Self-Contained Self-Rescuer Usage – Instructor’s
Copy. US Department of Health and Human Services, Centres for Disease Control and Prevention, National Institute for Safety
and Health Bulletin, October 1999.
Conti, R.S.; Linda L. Chasko, L.L; Wiehagen, W.J. & Lazzara, C.P. (2005). Fire response preparedness for underground mines.
Department of Health and Human Services. Information Circular/2005.
Haas, E.M.; Hoebbel, C.L. and Rost, K.A. (2014). An Analysis of Trainers’ Perspectives within an Ecological Framework: Factors
that Influence Mine Safety Training Processes. Science Direct, Safety and Health at Work 5 (2014) 118-124.
Kowalski, K.M., Mallett, L.G., Brnich, M.J. Jr., (1994). Leadership Characteristics in escape from Three Underground Mine Fires.
Mining Health and Safety Research Center.
Kowalski-Trakofler, K.M., Vaught, C., Brnich M.J. Jr., and Jansky, J.H. (2010). A Study of First Moments in Underground Mine
Emergency Response. Centre for Disease Control and Prevention, NIOSH Mining.
Minerals Council South Africa. (2018). Facts and Figures Pocket Book.
Mine Health and Safety Act, 1996 (Act No 29 of 1996). South African Department of Mineral Resources. Government Gazette 17725,
Notice No. R. 92, dated 16 January 1997.
Mine Health and Safety Inspectorate. (2011). Guidelines for the compilation of a mandatory code of practice for emergency prepar-
edness and response. Department of Mineral Resources.
Mine Health and Safety Act, 1996 (Act No 29 of 1996). South African Department of Mineral Resources. Government Gazette 17725,
Notice No. R. 92, dated 16 January 1997.
Mine Health and Safety Inspectorate. (2016). Guideline for the compilation of a mandatory code of practice for the prevention of fires
at mines (DMR 16/2/2/2-B1). Department of Mineral Resources.
Mine Health and Safety Inspectorate. (2011). Guidelines for the compilation of a mandatory code of practice for emergency prepar-
edness and response. Department of Mineral Resources.
Pearton, J.P.; Marx, W.M. & Willemsen, C.A. (2018). Fire engineering fit for the mining industry. Proceedings ‘Doing more with
less: Challenges for the Ventilation Professional’, The Mine Ventilation Society of South Africa, 2018 Conference.
Skeepers, N.C. and Mbohwa, C. (2015). A Study on the leadership behaviour, safety leadership and safety performance in the Con-
struction industry of South Africa. Proceeding of Industrial Engineering and Service Science.
114
Maximising spot cooling potential when utilising varying mine service-

water as heat sink
Dr. PvZ Venter & Prof. M. van Eldik

North-West University, South Africa and M-Tech Industrial (Pty) Ltd
J van der Merwe

M-Tech Industrial (Pty) Ltd
ABSTRACT: In recent times the health and safety of the mining workforce has been getting more focused
at-tention. Due to the hazardous and volatile underground environment, a better understanding was required
of the effects of environmental factors on health and productivity. The hazards are further compounded by
the expanding underground network and mines operating deeper every year. The increasing complexity of
under-ground works and rise in ambient temperatures, compounded by increased VRTs at depth, create
unique environmental and ventilation challenges. Effective and reliable ventilation and air cooling are es-
sential in providing a safe, productive workplace underground.
To keep operations economically sustainable, there is an increased focus on spot cooling methods in pro-
duction areas, thus eliminating unnecessary air cooling of non-operational areas.
Traditional spot cooling is achieved through an air-to-water heat exchanger (also known as a cooling car
(ACC)) using chilled water as heat sink in the area to be cooled. The cooling potential is however a function
of both the water flow rate and supply temperature. Therefore, the water temperature needs to be much lower
than that of the air to achieve any cooling. For vast underground mining operations this can only be achieved
via central cooling infrastructure consisting of either surface or underground fridge plants and a chilled water
recirculation system.
Another spot cooling method implemented, makes use of a mobile vapour compression refrigeration unit,
known as the Air-Cooling Unit (ACU), where water is also utilised as a heat sink. An attractive attribute of
the ACU is that it can deliver cooling over a large heat sink range, thus eliminating or reducing the central
cooling infrastructure requirements. The ACU can therefore cool air to temperatures lower than that of the
in-coming cooling water, which is not achievable with a conventional cooling car.
A further advantage is that an ACU can operate at a high energy efficiency level, therefore reducing opera-
tional costs. To ensure asset health, favourable operating conditions are required that include adequate supply
water flows. However, a mining environment comprises multiple water users at any single time, and it is
there-fore typically not possible to ensure continuous sufficient water flow availability twenty-four seven.
This results in lower ACU operational efficiency, and therefore higher operational costs.
This paper demonstrates the importance, and therefore, operational advantage when an ACU is used where
mine service-water is utilised as a heat sink. Furthermore, operational constraints where service-water is
avail-able in varying quantities, which results in ineffective air-cooling operations, both in terms of kilowatt
hour cooling provided as well as electrical costs per kilowatt hour of cooling provided are addressed through
im-proved compressor capacity control.
1. BACKGROUND
The mining industry makes use of underground air cooling and ventilation to provide a safe working environ-
ment for their workforce. It is not uncommon for underground air temperatures to exceed 35⁰C dry bulb with
115
high relative humidity percentages. These conditions are hazardous for humans to work in, and potentially fatal
without sufficient air conditioning. With an expanding underground network, working in remote areas, it is
sometimes very difficult to provide acceptable conditions without spot cooling, due to the excessive costs of a
centralised cooling system with its vast piping network, both in terms of capex and operational expenditures.
Spot cooling is a well-known concept where localised cooling is done only in a specific area, allowing personnel
to perform their work. As a result, areas with limited or no activities are not cooled, allowing for energy efficient
mining operations (Greyling, 2008 and van Eldik, 2007).
A traditional method of underground spot cooling is through an air-to-water heat exchanger, or air-cooling car
(ACC). Chilled water from either a surface or underground fridge plant is pumped to the designated area where
the air is cooled through an indirect heat exchanger. The water is therefore utilised as a heat sink medium to
transfer the thermal energy away from the area being spot cooled. The cooling potential of this traditional
approach is however a function of both the water flow rate and temperature, where the temperature is a lower
limit, i.e., air cannot be cooled below this temperature. Therefore, the water temperature needs to be substan-
tially lower than the air to provide effective cooling. This cannot always be guaranteed for underground mining
operations due to the vast depths and distances to remote working areas in modern mining operations and insu-
lation difficulties (Potgieter & van Eldik, 2017).
Another spot cooling method implemented makes use of a mobile vapour compression refrigeration unit, re-
ferred to as an Air-Cooling Unit (ACU), where water is also utilised as a heat sink. An attractive attribute the
vapour compression refrigeration cycle is that cooling can be deliver over a significantly large heat sink range.
An ACU can cool air to temperatures lower than that of the incoming cooling water, which is not achievable
with a conventional ACCs, and furthermore, eliminates or reduces the central cooling infrastructure require-
ments (Greyling, 2008).
However, the efficiency of ACU operations depends on the water availability. Within a varying service-water
availability environment, both the kilowatt hours air cooling delivered as well as the cost of operations can be
negatively influenced.
This paper demonstrates the importance, and therefore, operational advantage when an ACU, as opposed to an
ACC, is used where mine service-water is utilised as a heat sink. The ACC’s ability to perform air-cooling
while utilising mine service-water as heat sink is compared to that of an ACU. This includes water temperatures
where the ACC fails to deliver a safe working environment. Furthermore, the operational constraint of varying
service-water availability is addressed to maximise spot cooling potential. Improved screw compressor control
is incorporated and demonstrated through a case study, where spot cooling is maximised through increased
kilowatt air cooling provided, together with lower operational unit costs.
Note that for the remainder of this paper, mine service-water will only be referred to as water.
2. SPOT COOLING: A COMPARISON OF TECHNOLOGIES
Apart from equipment capacity sizing, the two largest influencing factors of underground spot cooling capaci-
ties are the supply water mass flow rate and temperature. The water mass flow rate is an indication of the ability
to transport extracted heat away from the area being cooled. The supply water temperature is an indication of
the ability to absorb heat. This difference between the two technologies follows.
A cooling car operates on the principal of direct heat extraction from a warmer (air) to a colder (water) medium.
Warm air is forced to flow over a heat exchanger with water inside finned tubes, where the fluids are thus not
in direct contact. If the cooling water temperatures are by some margin lower than that of the air, the cooling of
air will be achieved. For a cooling car to be effective this temperature difference, or the pinch, must be as small
as possible to increase the effectiveness thereof. Alfa Laval (Pty) Ltd
116
(https://www.alfalaval.com/microsites/increase-efficiency/products/compact-heat-exchangers/) claims that its

compact heat exchangers can transfer heat for a pinch as low as 2°C. This, however, will typically not be pos-
sible within a dusty underground mining environment.
The interpretation of a pinch is as follows; state that air must be cooled down to at least 25°C, via a cooling car
and that a pinch of 2°C is possible. For air cooling to occur, the water that enters the cooling car may not exceed
a maximum temperature of 23°C. Higher water temperatures will still allow for air cooling, only to a minimum
of 2°C above the available water temperature. Therefore, if the available water temperature is 30°C, the air can
only be cooled down to 32°C.
Figure 1: ACC cooling rates at water temperatures and - flow rates for dry-bulb air temperatures at 80% rela-
tive humidity.
The ACU evaluated can provide a rated cooling of 250 kW, i.e., at full capacity 250 kWh of heat can be ex-
tracted from the air in one hour for dry-bulb temperatures up to 40°C and a relative humidity of 100%. The
paper will therefore use this 250 kW as a basis, and for a comparison use a 250 kW ACC with a pinch of 2°C.
Assume now that air at 36°C (dry-bulb) and 80% humidity needs to be cooled. Figure 1 demonstrates the air-
cooling rates that are possible for different water temperatures over a range of mass flows when a cooling car
is used with a pinch of 2°C. Note that when water is at 20°C, approximately 5.0 l/s is required to deliver 250
kW air cooling; at 25°C 8.0 l/s is required and 30°C can only deliver 100 kW air cooling. Further note that no
cooling is possible when water flow rates are at 35°C.
Figure 2 provides the air cooling rates for an ACU under the same water flow conditions as in Figure 1. It can
be seen in Figure 2 that at 20°C, the ACU provides less cooling per water flow rate; however, slightly better
cooling rates are obtained at 25°C. Note how the full capacity cooling is still achieved when water is at 30°C
and almost 90% (220 kW) is possible at 35°C (dry-bulb).
117
Figure 2: ACU air cooling rates at water temperatures and flow rates for dry-bulb air temperatures at 80%
relative humidity.
From Figures 1 and 2 it is evident how the ACU outperforms the traditional ACC under high water temperature
conditions. Table 1 provides the minimum dry-bulb air temperatures that can be achieved by both ACC and
ACU under the water temperatures depicted in Figures 1 and 2. Note, even though the cooling car can provide
approximately 40% (100 kW) air cooling for 30°C water, the minimum achievable temperature is only 28°C
(dry-bulb), i.e., a pinch of 2°C. The ACU, however, can provide 23°C (dry-bulb) air temperatures for up to
30°C water temperatures under 35°C (dry-bulb) ambient conditions. Water at 35°C can achieve air cooling
down to 24.2°C (dry-bulb).
Table 1: Minimum achievable air-cooling temperatures for ambient air at 35°C (dry-bulb).
Water temperature (°C) ACC (dry-bulb) ACU (dry-bulb)
20°C 22.0°C 23.0°C
25°C 23.0°C 23.0°C
30°C 28.0°C 23.0°C
35°C 35.0°C 24.2°C
Note from Figure 2 that the ACU only provides air cooling from 2.0 l/s and higher, whereas it can be seen in
Figure 1 that an ACC can provide air cooling for any water flow rate above zero.
3. OPERATIONAL CHALLENGES FACED BY THE ACU
3.1 Real life case study comparison
This section provides two real-life case studies, i.e., Case study I and Case study II. Case study I demonstrates
unique challenges that are faced by an ACU that operates under varying water flow availability. Case study II
demonstrates ACU operations where improved compressor capacity control is implemented. From Case study
II, the inefficient operations from Case study I can be understood, as well as the operational improvements
under improved compressor capacity control. These operational improvements include increased air cooling
delivered in terms of kilowatt hours, at lower operational costs.
118
3.2 Case study I: four step control compressor operations
Figure 3: Supply water flow rate over time for Case study I
The water flow profile of Case study I is depicted in Figure 3, where water temperatures varied between 20°C
and 22°C. This is a typical, and therefore, not uncommon varying water profile that is experienced by ACUs
during operation. From Figure 2 it is both evident that different water flow rates result in different air-cooling
rates, and that the unit cannot operate when water flow drops below 2.0 l/s. For water flows below 2.0 l/s the
ACU must shut down to protect both the compressor and its motor. This is due to an excessive increase in
discharge pressure resulting from the inability of insufficient water flows to remove required heat from the
condenser. During these periods of unit shut down no air cooling can be provided, which is a concern for
continuous mining operations.
Important to note is that in Case study I, the compressor functions under four step-control operations, i.e., 25%,
50%, 75% and 100% of full capacity. As a result, when water flows vary as seen in Figure 3 the compressor
settings will change to try and keep the unit operational but prevents the cycle to operate optimally, and at times
the unit will shut down when operations is lower than the 25% setting. Due to the step control for certain periods
the unit performs at lower efficiencies, so that either less air cooling is provided, or equivalent cooling at higher
operational costs are experienced.
Figure 4 displays the electrical current following the water flow profile depicted in Figure 3. Although the water
flows fell below 2.0 l/s at times as can be seen in Figure 3, the non-operational instances do not align with only
flows below 2.0 l/s. These events are typically when low, but above 2.0 l/s water flows reduced too fast, so
residual heat built up in the vapour compression cycle, even though the compressor was at stepped down to
25% capacity. As a result, unnecessary shut down events occurred due to low flow water variability.
Table 2 provides the cooling COP of the ACU operating under varying water flow conditions, as depicted in
Figure 3. Note that for steady state design conditions the cooling COP of an ACU that operates between 5.0 l/s
and 6.0 l/s should be 3.3 for the current temperature range (20°C to 22°C), as opposed to 2.5 in Table 2.
A solution to improve the performance is for the supply water flow rate to be stable within the design specifi-
cation of the ACU. Water, however, is a scarce resource, and furthermore, the availability thereof for under-
ground mining depends on numerous factors, which include infrastructure and shift operations. It is typically
not viable to request stable and sufficient water flow availability as it is difficult for the mine to control. There-
fore, an alternative solution is required for the ACU to stay operational under varying water flow availability.
119
Figure 4: Electrical current over time for Case study I.
Table 2: Cooling COP at different water flow rate intervals for Case study I.
Water flow rate [l/s] COP
2 l/s to 3 l/s 2.0
3 l/s to 4 l/s 2.1
4 l/s to 5 l/s 2.3
5 l/s to 6 l/s 2.5
3.3 Case study II: improved stepless compressor capacity control
There can be stepless capacity control on the screw compressor slide valve mechanism where the discharge
pressure is controlled within a small interval, i.e., near constant, so that the compressor step-up or step-down is
automated, and not controlled in fixed steps as in the previous section. Case study II is for an ACU where the
screw compressor operates with a stepless capacity controller.
Figure 5 displays water flows for Case Study II’s ACU and Figure 6 its electrical currents. Further note that
flow variability is more prominent in Figure 5 when compared with Figure 3. The expectation would be that
cooling COPs based on ACU operations without a stepless capacity controller under Figure 5 will not be an
improvement when compared to those provided in Table 2.
However, Figure 6 depicts the electrical currents with the varying water flows of Figure 5. Note the less ‘scat-
tering’ of data points in Figure 6 as opposed to Figure 4. This is an indication of fewer shutdowns that were
experienced by the ACU in Case study II with higher water flow variability as opposed to Case study I. The
cooling COPs for Case study II’s ACU are provided in Table 3.
Figure 5: Supply water flow rate over time for Case study II
120
Note from Table 3 how the cooling COPs, and therefore, the cycle efficiency is constant over the range of water
flows, despite the variability thereof. This is due to the stepless capacity controller that allows the compressor
to operate at a near constant discharge pressure. As a result, the unit operating in Case study II operates at more
efficient levels than the unit from Case study I. Further note that the water profile in this case never exceeded
5.0 l/s, so that cooling COP information cannot be provided for this instance.
Figure 6: Electrical current over time for Case study II.
Table 3: Cooling COP at different water flow rate intervals for Case study II.
Water flow rate [l/s] COP
2 l/s to 3 l/s 3.3
3 l/s to 4 l/s 3.3
4 l/s to 5 l/s 3.3
3.4 Kilowatt hour cooling and electrical cost comparison between Case studies I and II
For Case study I, an average hourly air cooling of 196.5 kWh was achieved, while the screw compressor power
usage was 93.3 kWh (COP=2.1). It follows from a developed simulation model that if the unit was to be oper-
ated with a stepless capacity controller under the same water profile, an average hourly air cooling of 223.4
kWh could possibly have realised at an electrical power input of 67.3 kWh (COP=3.3). Therefore, the capacity
controller would have resulted in a 13.7% increase in the air-cooling rate, together with a 17.8% reduction in
electrical power consumption per kilowatt hour of cooling delivered.
Note that for power consumption only the refrigeration cycle was considered, which excluded all auxiliary
operational costs that include the air fan.
4. SUMMARY AND CONCLUDING REMARKS
This paper discusses the ability of a spot cooling ACU to provide a safe working environment for underground
mining where centralised air conditioning is either not possible or practical, or where water temperatures are
too high for conventional ACC utilisation. Operational challenges in terms of varying water flows were further
discussed as well as the reduced operational efficiency and less air cooling that may occur.
A compressor capacity controller was implemented as a solution that allows the ACU to operate at higher air-
cooling levels, together with increased energy efficiency operations. A simulation based on data analysis mod-
els indicated that an ACU operating with a stepless capacity controller for the compressor could have provided
13.7% more air-cooling, at an electrical cost reduction of 17.8% per kilowatt hour cooling provided compared
to a step control approach.
121
REFERENCES
1. Greyling, J. 2008. Techno-economic application of modular air cooling units for deep level mining at Mponeng.
Potchefstroom: NWU. (Dissertation - M.Eng.).
2. Potgieter, R. & van Eldik, M. 2017. Mine ventilation system optimisation considering optimal energy, health and
safety. 13th Interna-tional Mine Ventilation Congress. South Africa.
3. Van Eldik, M. 2007. An investigation into the DSM and energy efficiency potential of a modular air cooling unit
applied in the South African mining industry. Potchefstroom: NWU. (Thesis – PhD).
122
Interactive Training Model
FT Maimane
AngloAmerican, Amandelbult Complex
ABSTRACT: Emergency preparedness is one of the most imperative subjects in health, safety, and Business
sustainability. Therefore, it is of utmost importance that our employees underground are aligned and assertive
in events that call for an emergency remedial response. It is further highlighted that a training model is highly
evident to be established in order to ensure users are conversant with the necessary steps to take with the aim
to evacuate in the event of a fire, flammable gas intersection, ventilation failure, power failure, life-
threatening events, etc.
The Model is built in different communicative platforms to interact with employees on what to do in scenarios
whereby a life-threatening event occurs or has a likelihood to occur.
· Interactive animation
· Animation explanatory
· Virtual Reality
This will provide the ability of the user to have a more virtual reality facilitating method that will interpret the
different underground sequences of events which therefore require an agile response in order to protect the
health and safety of mine employees. The whole concept is to be uniformly introduced to address all the controls
required to be put in place in case of an emergency and thoroughly comprehend vital/critical mine ventilation
procedures to retain all the set of skills that have been acquired in the training duration
1. INTRODUCTION
The ventilation training model is built to enable employee participation in the mining industry. This then
accelerates more employees to engage with the technical aspects of fluid dynamics, fire prevention and
emergency preparedness with the aid of computer graphics technology. This enables us to construct a
remarkable variety of digital images and displays the right conditions to effectively enrich the level of
education. Computerized graphics are derived to enable people to deal with information easily and is
successfully developed to facilitate learning for over 20 years since its development ( (Bricken, 1991). It is
important to note, learning comes with experience and through digitalised training we then allow mine
inhabitants to experience incidents where there is a defect in mine ventilation circuits, then the user has the
opportunity to derive solutions from the program.it is regarded that, normal learning human is considered to
utilise his or her own efforts to attain educational goals with the aim to gain a positive outcome.
When we familiarize ourselves with digitalized training, we create purposeful memorial transactions and
effective brain domains to allow natural interaction with the information provided. The skill that digitalised
learning provides a novice is so compatible to enable them to showcase what they have learned in the physical
realm through further vulnerary programming of the mind. Virtual Reality is a subject matter which enriches
the skillset which cannot be taught in the physical world to be easily comprehended in front of a screen. The
123
user utilises a high proportion of their brain capacity to easily embed the administrated information, enabling a
more subject matter competency and enables a higher confidence level on the outcome.
Amandeldelbult complex came with the initiative to roll out a computerized programme that comprises a build-
up process of ventilation standards with the aim to make effective training easier and interactive.
2. HISTORICAL EVENTS
2.1 South African mine disasters
It is identified from hysterical occurrences how evident it would’ve prevailed a different outcome with regards
to preparedness of employees underground during events which took place in the past. Table 1, illustrates these
events which took place in the South African mining industry limited to major fatalities from the 20th century.
These disasters witness how important it is that all mine employee categories have to form an integral part in
the integrated training model, with aim to having a high confidence level with regards to necessary steps to take
if a life-threatening event is likely to occur. The chart following shows 72% of mine disasters was mainly due
to fires and explosions, which is then ideal to realize that we cannot become complacent in this paradigm, but
to broaden the comprehension of underground personnel on a more visual perception to create an ease of con-
fidence in decision making.
Table 2: Historical events in the mining industry
Year Fatalities Energy Mine Location Summary

A miner relighting his lamp while checking for methane gas, ignited
1944 57 Fires and Explosions Hlobane Colliery the methane gas and caused a massive explosion.
1960 435 Fall Of ground Coalbrok colliery Underground pillars supporting the roof collapse
The methane gas and dust explosion tore through the mine when a
1983 64 Fires and Explosions Hlobane Colliery flame-proof electrical drilling machine ignited the gas
1985 33 Fires and Explosions Middelbult Colliery methane gas explosion
acetylene cylinder ignited the polyurethane foam that covered the
1986 177 Fires and Explosions Kinross Gold mine side-walls of the mining tunnel
1987 34 Fires and Explosions Ermelo Mine methane gas explosion
1987 62 Fires and Explosions St Helena Gold Mine Methane gas explosion
1990 21 Fires and Explosions Vaal Reefs
A major coal-dust explosion was caused by a small volume of methane
1993 53 Fires and Explosions Middelbult Colliery gas that came into contact with a source of heat
Rail bound underground locomotive fell down a shaft and landed onto a cage
1995 104 equipment Vaal Reefs loaded with night shift mine workers coming to surface
1996 20 Inundation Rovic diamonds Inundation
124
MINE ACCIDENTS AND DISASTORS

Inundation
Rail bound
equipment
Fall Of
ground
Fires and
Explosions
Emergency preparedness and Fire prevention are one of the most important topics in the mining industry to
ensure certainty to sustain the lives of employees working in an underground mine. It is imperative that em-
ployees underground while working in the mine are cognizant of what measures or steps are required to take
place to ensure there is no interruption in correspondence to the health and safety of each employee.
Over the past events, it has been provided that there’s a requirement for an intervention in our mining industry
to improve our confidence level in psychological safety and work-related safety of our mining employees un-
derground, which stimulates their competency on what remedial steps they would need to take upholding their
well-being (Figure 1). We looked into our hierarchy of controls which administrative training and communica-
tion became a good strategy to enable our underground employees to be conversant on different ventilation
critical controls and steps to take during emergency preparedness events.
To make it more embedded on the remedial requirement to improve the industry’s training systems, one of the
most recent events on the 15 of July 2018. Six mine workers were found deceased in a refuge chamber in a
result to fire from a conveyor belt. One of the most critical issues highlighted was that, the employees could not
identify the issues involved in the refuge chamber on their daily routine checks and it was also highlighted that
the refuge bay was located 1400m away from its area of influence. (Hermann Fasching, 2018 ). Such historical
events make transparent on the importance of how mine employees need to be conversant with emergency
preparedness constraints and ventilation practices to enable everyone to be proactive if a life-threatening situa-
tion arises.
2.2 Current mitigation controls (training)
The current systematic approach in the mining industry to improve most of the compliance criteria. This is
achieved by providing a retraining schedule to employees who have deviated from any ventilation critical
controls which participants read through the provisioned training material and write a written exam, then re-
turn back to normal operations. This has a linkage to the approach provisioned in order to comply with ISO
45001:2018 standards, the following is considered:
1. Induction
An annual induction is conducted in the criteria to assure legal compliance by assuring employees are
trained on their legal and mandatory obligations, this principle includes Gas detection training and
underground fire drills on induction courses in jurisdiction where participation of employees is required
125
on the class facilitated programme in place to meet with the relevant human resource development
objectives derived through a communication base training with a competent facilitator.
2. Communication – It is agreed in general that most problems in emergency response come from a
breakdown in communication and misunderstanding of information. In the study presented by Mallet,
it considers that initial information is good only when it is being received, while critical information
must be accurate, precise and is formed from a trusted source (Kathleen M, 2006). In the topic of
emergency preparedness response, it is important to make informed decisions, as we regard employees’
need to be ready to participate in any form of emergency procedure they encounter, more especially
related to fire-related incidents.
2.3 Legislation
MHSA 10.1(a), (b) As far as reasonably practicable, every employer must-0
(a) provide employees with any information, instruction, training or supervision that is necessary to enable
them to perform their work safely and without risk to health; and
(b) ensure that every employee becomes familiar with work-related hazards and risks and the measures that
must be taken to eliminate, control and minimize those hazards and risks.
The aim is to ensure the employees working at the mine are conversant of ventilation legal standards includ-
ing the following procedures guided by DMRE guidelines. The following documents were integrated in refer-
ence to the training:
• Prevention of Fires at Mines

• The prevention of Flammable Gas Explosions in Mines other than Coal Mines.
• Early warning system for fires and toxic releases underground
• Emergency preparedness and response
Embedding these procedures with an integrated training model will relatively increase the employer’s confi-
dence level in the employee’s preparedness on the event of a life-threatening occurrence relating to fires and
explosions underground.
3. TRAINING MODEL
3.1 Animated training model (information phase - phase 1)
The purpose of the animated training will be outlined to enable the user to have a more interactive session
comprising of a question and answer criteria that intends to highlight important topics in mine ventilation
engineering. These topics are currently facilitating several topics including mining ventilation, evacuation
procedures and detection devices.
126
Figure 3: panel control modelling
The above figure gives a snapshot of how the model interacts with the user when they are in training (Phase 1),
the user will be enabled to gain a visual illustration of what will be the change in ventilation flow if there is a
non-adherence to the recommended infrastructure. Figure 1 point (A) makes the learner cognizant about the
face conditions if there are controls put in place to direct air to the advancing face.
The industry is currently led by facilitator training, which is currently based on PowerPoint presentation and
estimated to 1-2 hours session to complete. The user base literacy level of the current programme is at a Matric
(Grade 12) education NQF level 4, aiming the interactive training to have the following outcomes:
1. Interactive – ease of interpretation

2. Easy to consolidate the information – User can absorb information from their language preference
3. User friendly – direct communication platform and data captured
4. Covers the latest ventilation standards- Ease of access to update if there is a change in the procedures
5. Explain abstract ventilation concepts to a new user- technical information which will be easy to retain as they
train through animation.
A CBT (Computer Based Tutorial) animation is used to illustrate points of interest in mine ventilation engi-
neering. The animated programming is selected to demonstrate various functions in mine ventilation engineer-
ing field by enabling the user to understand more of the theory while it is interpreted with visual motions under
their control for better accuracy in training.
Figure 4: illustration of column distance tends velocity monitoring

127
In the process of deploying the software platform in the training process, it is beneficiary for the execution and
assessment of the proposed course material. The course material is a web-based training, so the training and
assessment products will have full access over the mine operation network, not just in the Training Center but
all employees requiring further clarity in the complex will have ease of access for information purposes. The
illustration below forms part of the training which then provides steps to be taken by employees in case of
emergency and in occurrences where an employee is deemed required to utilize a life sustainable refuge bay in
conjunction with donning a self-contained rescue pack. Note, this only form part of the first phase of the training
which composes of 2 phases until further integration within the Ventilation fraternity.
Figure 5: Employees safely evacuated to a refuge bay (Animation)
3.2 Virtual Reality (Interactive phase - Phase 2)
The virtual reality model will be enrolled on completion of phase 2. The use of 2 phases in implementation will
assist the user to be comfortable with the covered topics. It utilizes realistic simulations that provides a better
feel of how the environment will be in different situations pertaining to emergency preparedness and fire pre-
vention. It has been evident that virtual training has a better reduced cost constraint and greater confidence in
safety.
i. Location and freedom

ii. Time flexibility
iii. High knowledge retention
iv. Less time consuming
v. Immediate evaluation
Virtual Training
Pros Cons
Time and location freedom Technical issues, easy to fix in most cases
Convenience
Cost effective
Improved technical knowledge
Access to worldwide audience
Hands on participation
128
Traditional in person course

Pros Cons
Face to face interaction Transportation and delay
Maintenance in participative focus No hands-on learning experience
Time consuming
Limitation of advanced technology
(Gevorgya, 2023)
The user will input their information and be provided with various courses in emergency preparedness. This
allows a better interaction with the environment and what they will do in the event of a life-threatening occur-
rence. The virtual training will further enhance information retention and improve user proactive capabilities.
Figure 6: Anglo American Virtual reality scope of work
The use of virtual reality in a workplace environment isolates the employee from the ambient environment, its
uniqueness prioritises the user and enables them to gain hands-on participation. This will neutralize the
employee from any movement around them and place them in the following parameters but not limited to:
i. Unventilated locations (Flushing procedure)

ii. Flammable gas detection
iii. Power failure
iv. Ignition from a source
v. High risk ventilation location
vi. Potential fire risk
vii. Donning of Self-Contained rescuer
The motive behind the concept was to look at the following criteria for decision making in an event where
remedial action is required:
i) Situational – Employees need to understand the type of problem they are dealing with and its
severity. They need to be aware of who are the people involved as role players.
ii) Organizational – understanding of each individual’s role in the event of an emergency evacuation.
129
iii) Cognitive – How important is the source of communication and information, A full interpretation
of information accuracy and adequacy can be evidently controlled.
iv) Autonomic – Good training effectively results in good decision making (Acumen). When one is
effectively trained this impacts instinct to reduce numerous decisions to be made.
v) Physiological – having inadequate understanding in a situation may lead to stress which evidently
leads toad decisions. (Kathleen M, 2006)
The virtual reality technology brings in an experience criterion which maximizes the capacity of employees
underground to make decisions remedially as they have experienced the event before it has occurred.
4. CONCLUSIONS
Utilization of artificial intelligence/ digital training integrated with ventilation engineering and emergency pre-
paredness training has the capabilities to ensure that employee performance can improve or incline better en-
gagement in competent. The training will also allow limited literacy level users to gain a better perception to
improve accuracy and knowledge retention, this is in comparison to reading a guideline or training material
which provide minimal potential of comprehending information. Interactive training has an exponential capa-
bility to improve engagement, knowledge and experience. Objectivism and constructivism represent concepts
for learning or thinking, this enables the mental activities to represent occurrences in the real world using com-
puterised VR tools. People become more participative in unique learning environments (Bricken, 1991) and
create different interpretations of the attained information through their genuine mental capacity. Animated
digital training will help us engage more effectively and will enable the industry to acknowledge how much
understanding the receiver has to actively be prepared for any form of dangerous occurrence. This programme
has a substantial potential to maximise safety and health in any mining environment.
130
5. REFERENCES
1. Bricken, M., 1991. Virtual Reality learning environments: Potentials and Challenges. HUman Interface Technol-
ogy Lab , 25(Number 3), p. 7.
2. Cindy, B., 2022. Rock Engineering training programme [Interview] (17 August 2022).
3. Gevorgya, N., 2023. Utech. [Online]
Available at: https://uteach.io/articles/virtual-training-effectiveness
[Accessed 17 June 2023].
4. Hermann Fasching, 2018 . South African copper mine conveyor fire. REsources safety and health, 1(1), p. 2.
5. Kathleen M, K. T. C. V. a. J. H. J., 2006. A study of first moments in underground mine emergancy response.
6. Pepper, l., 2020. Whatech. [Online]
Available at: https://www.whatech.com/og/markets-research/it/638316-global-immersive-virtual-reality-market-
study-rivalry-landscape-regional-analysis-and-forecast-up-to-2025
[Accessed 22 May 2023].
7. Phefo, O., 2022. NW Rustenburg tripatate. Department of Mineral resources and energy , Volume 1, p. 8.
131

MVSSA Conference 22 23 June 2023 Proceedings

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MVSSA Conference 22 23 June 2023 Proceedings

Uploaded by

Copyright:

Available Formats

THE

MINE VENTILATION SOCIETY

2023 Annual Conference

22 and 23 June 2023

© 2023 The Mine Ventilation Society of South Africa

The Mine Ventilation Society of South Africa

© The Mine Ventilation Society of South Africa, 2023

© 2023 The Mine Ventilation Society of South Africa

© 2023 The Mine Ventilation Society of South Africa

The manuscript review committee consisted of the following members:

The organizing committee consisted of the following members:

© 2023 The Mine Ventilation Society of South Africa

Planning and designing an efficient mine ventilation network by FS Bergh 1

Women and ergonomic risk factors in underground coal mining. by OS Mokwena 33

Is carbon offset realistic for South African mining industry by Z. Jordaan 41

A review of the application of Continuous Real-Time Monitoring in Noise Risk Management by K. 69

Interactive Training Model by FT Maimane 123

© 2023 The Mine Ventilation Society of South Africa

Planning and designing an efficient mine ventilation network

© 2023 The Mine Ventilation Society of South Africa

Section 4 presents conclusions.

Section 5 focuses on recommendations.

Section 6 lists all sources used for information gathering.

© 2023 The Mine Ventilation Society of South Africa

© 2023 The Mine Ventilation Society of South Africa

Furthermore, the software to be utilised to identify bottlenecks , areas of excessive resistance

Another example is to investigate Ventilation on Demand technologies, where a tracking so-

2.3 Mine Ventilation Planning Throughout the Life of Mine

© 2023 The Mine Ventilation Society of South Africa

2.4 Synergies and Interaction Between Departments

© 2023 The Mine Ventilation Society of South Africa

3. RESULTS AND ANALYSIS

© 2023 The Mine Ventilation Society of South Africa

© 2023 The Mine Ventilation Society of South Africa

© 2023 The Mine Ventilation Society of South Africa

Tailoring plant engineering to fast-track supply and construction of re-

1.1 Existing surface refrigeration plant

© 2023 The Mine Ventilation Society of South Africa

Table 1. Original thermal design parameters for the pre-cooling tower

© 2023 The Mine Ventilation Society of South Africa

1.2 New Cooling System Thermal Performance

Table 2. Thermal design parameters for the additional cooling system

© 2023 The Mine Ventilation Society of South Africa

1.3 New Cooling System Design Concept

Figure 1: Process Flow Diagram for Additional Cooling System

The key areas to consider in order to fulfil this concept included:

2. NEW COOLING SYSTEM DESIGN FEATURES

2.1 In-line construction

© 2023 The Mine Ventilation Society of South Africa

2.2 Service water isolation from refrigeration machines

Figure 2: Plate heat exchanger bank with upstream strainer

2.3 Condenser cooling tower selection

© 2023 The Mine Ventilation Society of South Africa

Figure 3: Factory-assembled package cooling towers

2.4 Substation construction

© 2023 The Mine Ventilation Society of South Africa

Figure 4: Aerial view of site during construction

© 2023 The Mine Ventilation Society of South Africa

Preparing a mines ventilation network to support expansion of the

© 2023 The Mine Ventilation Society of South Africa

2. VENTILATION MODELLING: THE CALIBRATED MODEL

© 2023 The Mine Ventilation Society of South Africa

Figure 1: The Calibrated Model

2.1 Analysis of the calibrated model