Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Available online at www.sciencedirect.

com

ScienceDirect
Available
Available online
online at
at www.sciencedirect.com
www.sciencedirect.com
Procedia Computer Science 00 (2019) 000–000

ScienceDirect
ScienceDirect www.elsevier.com/locate/procedia

ProcediaComputer
Procedia ComputerScience
Science00
148 (2019)
(2019) 226–235
000–000
www.elsevier.com/locate/procedia

Second International Conference on Intelligent Computing in Data Sciences (ICDS 2018)

Risk Assessment of Maintenance activities using Fuzzy Logic


Second International Conference on Intelligent Computing in Data Sciences (ICDS 2018)
Maryam Gallaba,c*, Hafida Bouloizb, Youssef Lamrani Alaouic, Mohamed Tkiouatc
Risk Assessment of Maintenance activities using Fuzzy Logic
SMARTILAB-EMSI, Moroccan School of Engineering Sciences, EMSI, Rabat, Morocco
a
b
Industrial Engineering Department, National School of Applied Sciences, ENSA, UIZ, Agadir, Morocco
Maryam Gallab *, Hafida Bouloiz , Youssef Lamrani Alaoui , Mohamed Tkiouatc
c a,c
LERMA, Engineers’ Mohammadiab School, EMI, Mohamed V University, Rabat, c Morocco

a
SMARTILAB-EMSI, Moroccan School of Engineering Sciences, EMSI, Rabat, Morocco
b
Industrial Engineering Department, National School of Applied Sciences, ENSA, UIZ, Agadir, Morocco
c
LERMA, Engineers’ Mohammadia School, EMI, Mohamed V University, Rabat, Morocco

Abstract

This paper presents a new model to quantify the risks associated with maintenance activities by coupling the risk analysis method
with fuzzy logic. The goal is to calculate the qualitative values of the risk level. The quantification of the Risk Priority Number
Abstract
(RPN) is based on three parameters: frequency, detectability, and severity. However, its result is often subjective and does not
present the exact value. In addition, the combinations of different scores related to the three aforementioned parameters may have
the same
This papervalue of RPN
presents a neweven when
model to the importance
quantify of the
the risks risks is with
associated not the same; hence
maintenance the utility
activities of fuzzy logic.
by coupling A analysis
the risk model based on
method
the fuzzy
with fuzzysets theory
logic. is proposed
The goal for assessing
is to calculate the risk values
the qualitative level ofof maintenance
the risk level.failure scenarios in the
The quantification LPG
of the supply
Risk chain.
Priority Such
Number
approach
(RPN) allowson
is based choosing priority failures
three parameters: that affect
frequency, equipmentand
detectability, andseverity.
the whole supply chain
However, system.
its result is often subjective and does not
present the exact value. In addition, the combinations of different scores related to the three aforementioned parameters may have
© 2019
the sameThe Authors.
value of RPNPublished
even whenby the
Elsevier B.V. of the risks is not the same; hence the utility of fuzzy logic. A model based on
importance
Keywords:Maintenance, risk quantification, fuzzy logic, supply chain.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
the fuzzy sets theory is proposed for assessing the risk level of maintenance failure scenarios in the LPG supply chain. Such
Peer-review under responsibility of the scientific committee of the Second International Conference on Intelligent Computing in
approach allows choosing priority failures that affect equipment and the whole supply chain system.
Data Sciences (ICDS 2018).
1. Introduction
Keywords:Maintenance, risk quantification, fuzzy logic, supply chain.
Maintenance is considered as an indispensable activity within a company. The aim is to ensure the operation of the
industrial system and to keep the equipment in a satisfactory condition for the performance of the intended
1. Introduction
functions. Nevertheless, maintenance activities have been identified since a long time as potentially critical
Maintenance is considered as an indispensable activity within a company. The aim is to ensure the operation of the
industrial system and to keep the equipment in a satisfactory condition for the performance of the intended
functions. Nevertheless, maintenance activities have been identified since a long time as potentially critical
* Corresponding author.
E-mail address: meryam09@gmail.com / Tel: +212 6 11 83 18 05

* Corresponding author.
E-mail address: meryam09@gmail.com / Tel: +212 6 11 83 18 05
1877-0509 © 2019 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/)
Peer-review under responsibility of the scientific committee of the Second International Conference on Intelligent Computing in Data Sciences (ICDS 2018).

1877-0509 © 2019 The Authors. Published by Elsevier B.V.


1877-0509 © 2019 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
This is an openunder
Peer-review accessresponsibility
article under the ofCCtheBY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/)
scientific committee of the Second International Conference on Intelligent Computing in
Peer-review under responsibility
Data Sciences (ICDS 2018). of the scientific committee of the Second International Conference on Intelligent Computing in Data Sciences (ICDS 2018).
10.1016/j.procs.2019.01.065
Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235 227
Gallab et al./ Procedia Computer Science 00 (2019) 000–000 2

situations and present risks that can lead to serious accidents [1- 20]. For these reasons, risk analysis must be one of
the major challenges of each maintenance activity.

Several models have been proposed to analyze risk in maintenance activities. Among these models: FMEA (Failure
Mode Effect Analysis), HAZOP (HAZard and OPerability study), MOSAR-UML [21] ... To calculate the risk level,
these methods depend on three parameters: frequency, detectability, and severity. However, the result obtained is a
quantitative value that does not consider the uncertainties. The combination of different scores of these three
parameters may herd to the same risk level value (RPN) even when the importance of the risks involved is not the
same.

In the absence of a quantitative probability model, fuzzy logic can classify the key risks consistently, considering
both available data and expert opinions [22]. Shang in [22] adds that such model can help to identify the serious
risks and may include information about the causes of risk exposure or factors that have a significant impact on it.

The aim of this paper is to establish a model for risk quantification based on the fuzzy logic sets. The idea is to
calculate risk level or RPN values to achieve significant results. These results allow choosing priority failure
scenarios related to maintenance activities that need to be worked on. Indeed, the RPN calculation is based on three
parameters: severity, frequency, and detectability. Thereby the fuzzy sets are used to analyze these parameters and
their integration into risk quantification.

This paper is structured as follows: Section 2 proposes a risk quantification model using fuzzy logic. Section 3
presents an LPG supply chain as a case study in order to show the results. Finally, a conclusion is provided.

2. The proposed model: risk quantification using fuzzy logic

2.1. The methodology

Fuzzy logic is based on the theory of fuzzy sets developed by Zadeh [23]. It is a generalization of the classical set
theory. This technique provides flexibility for reasoning and considers inaccuracy, subjectivity, uncertainty, and
imprecision [24]. Fuzzy logic provides wide opportunities for working with imprecise linguistic data by defining
rules and membership functions in sets called “fuzzy sets” [25].
Fuzzy sets theory is applicable when assessing indicators for which there is no conventional model for estimating
and measuring or if the model is too complex. According to Zadeh [23, 26, 27], this theory is the most suitable
formalism to describe qualitatively the linguistic variables.
Indeed, fuzzy logic is applied for risk assessment and reliability [28-32]. The advantage of fuzzy theory
application for risk assessment lies in the fact that the resulting system assessment is qualitative and the ability to
operate with linguistic variables since some events cannot be described numerically. Whereas, fuzzy logic deals
with subjective, incomplete or unreliable knowledge bases.
Fuzzy logic is an appropriate tool that can determine easily and accurately the critical elements of the system. It
takes each risk factor level and evaluates them simultaneously to infer their joint contribution to operational risk
indicator. These can help in determining and implementing the correctives measures for reducing risks.
The fuzzy inference is a formulation process of the input data to an output data using fuzzy logic. It includes all
these episodes: Membership function, fuzzy logic operators and If-then rules [33] [36].
The choices made by the designer of a fuzzy system such as defining the membership functions and the decision
matrix are based mainly on the advice of the expert or statistical data [24]. A fuzzy logic system is established based
on the steps depicted in Fig. 1:
 Select the key indicators that affect the dependent variables;
 Create fuzzy sets for both independent and dependent variables and then specify the degree of truth that each
variable belongs to a certain fuzzy set using the membership functions;
 Establish the inference rules in the system;
228 Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235
Gallab et al./ Procedia Computer Science 00 (2019) 000–000 3

 Generate the output fuzzy set of the dependent variable based on the independent variables and the inference
rules, then a numerical value of the output fuzzy set is calculated by the defuzzification operation;
 Use the model results to make a decision.

Fuzzification •Input: Parameters

Fuzzy
Inference •Rules
System

Defuzzification •Output: RPN

Fig. 1. Overview diagram of a fuzzy system.

2.2. The proposed framework

We defined first the input, which is the three parameters: Severity, Frequency and Detectability. The values of
these three parameters are presented in the tables 1, 2 and 3. These tables represent the numerical values of each
parameter with the linguistic terms.
The severity is defined by fuzzy sets with the following terms: Very Low (VL), Low (L), Medium (M), High (H)
and Very High (VH). The term “VL” represents the moderate failures which almost do not influence the system
function, while the term “VH” describes the disastrous failures that cause great harm to the system, and serious
injuries. The fuzzy sets are associated with the corresponding class from 1 to 5 and are represented by a triangular
and trapezoidal membership functions. The fuzzy sets of the severity are shown in Fig. 2.

Table 1. Severity of failures scenarios.

Severity parameter

Rank 1 2 3 4 5

Value Moderate Serious Important Catastrophic Disastrous


Linguistic
VL L M H VH
Term

Fig. 2. Shape of severity variable.


Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235 229
Gallab et al./ Procedia Computer Science 00 (2019) 000–000 4

The frequency of failure occurrence can be determined by a set of five ranks of failure modes occurrence, which are: Very
Low (VL), Low (L), Medium (M), High (H) and Very High (VH). The very high frequency is the event that occurs several times,
while the fuzzy set called ‘‘very low’’ describes an extremely unlikely situation that can occur more rarely. The fuzzy sets with
failure occurrence ranking and the corresponding membership functions are shown in Fig. 3.

Table 2. Frequency of failure scenarios occurrence.


Frequency parameter
Rank 1 2 3 4 5
Event
possible but Event very Event Event Current
Value
extremely unlikely unlikely likely event
unlikely
Linguistic
VL L M H VH
Term

Fig. 3. Shape of frequency variable.

The detectability of failure can be easy, i.e. it appears to all staff; or very difficult to detect that means it requires
level sensors and detectors to be detected. Like the previous two parameters, the fuzzy sets for possibilities of failure
detection are formed with the following linguistic variables: Very Low (VL), Low (L), Medium (M), and High (H).
The fuzzy set ‘‘H” includes failures that need expertise. The events or failures which are easy to detect with anyone
on the staff are grouped in the fuzzy set ‘‘VL’’. Ranks of possible failure detection are given with a numeric value
from 1 to 4 and presented by a triangular and trapezoidal membership functions as depicted in Fig. 4.
Table 3. Detectability of failure scenarios.
Detectability parameter
Rank 1 2 3 4
Easily
Value Little sign No sign Expertise
detectable
Linguistic
VL L M H
Term

Fig. 4. Shape of Detectability variable.

Based on the membership functions of the input variables, 100 inference rules are generated; the RPN variable is
described in terms of the three risk parameters: severity (S), frequency (F), and detectability (D) (Fig. 5) using Table
4, 5, 6, and 7. These tables are obtained by the expert opinion based on their own analysis. Risk fuzzy sets are
presented in Fig. 6. To get a numerical value of the RPN related to each risk, defuzzification is done using the
230 Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235
Gallab et al./ Procedia Computer Science 00 (2019) 000–000 5

centroid method. For example: if the detection is very low (D = VL), the frequency is very low (F = VL) and the
severity is very low (S = VL), thus the RPN is very weak (RPN = VL).

Fig. 5. The fuzzy model.

Table 4. Fuzzy computation of the output RPN for D= VL.


Table 6. Fuzzy computation of the output RPN for D =M.
D= VL
D= M
F F
RPN
VL L M H VH RPN
VL L M H VH
VL VL VL VL VL VL
VL L L L M M
L VL VL VL L L
S L L L M M M
M VL L M M M
H L M M M H S M L M M H H
VH L M M H H H M M H H H

Table 5. Fuzzy computation of the output RPN for D = L. VH M H H H VH

D= L
Table 7: fuzzy computation of the output RPN for D =H.
F
RPN D= H
VL L M H VH
VL VL VL L L L F
RPN
L L L M M M VL L M H VH
S M L M M M H VL L L M M M
H M M H H H
L L M M M M
VH M H H H H
S M M M H H H
H M H H VH VH

VH H H VH VH VH
Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235 231
Gallab et al./ Procedia Computer Science 00 (2019) 000–000 6

Fig. 6. Risk fuzzy sets of risk level.

Based on the established set of inference rules (Table 4, 5, 6, and 7), the fuzzy logic inference provides the risk
level results according to any combination of the key risk parameters. These combinations are represented as a
surface plot and illustrate three-dimensional (3D) risk profiles (Fig. 7).

Fig. 7. Risk level as a combination of the different risk


parameters.
Gallab et al./ Procedia Computer Science 00 (2018) 000–000

232 Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235

3. Case study: LPG Supply Chain

The case study presented in this paper concentrates on the Liquefied Petrol Gas ‘LPG’ site that assures its supply
from suppliers and distributes LPG to their customers. The LPG unloading is done by the liquid/gas arm to fill the tank
using the compressor. The LPG loading is performed by the liquid arm attached to the tank using the pump to ensure
the filling in good conditions [20, 34, 35] as mentioned in Fig. 8.

Fig. 8. The representation of the selected equipment [20].

Several maintenance failure scenarios have been detected in this equipment: pumps, compressors, and liquid/gas
arm, which are presented in Table 8.
The risk level (criticality) or RPN is calculated based on the established inference rules (Fig.9), from the three
parameters (Severity, Occurrence, Detection) using the fuzzy logic inference system as shown in Fig. 10.

Fig.9. Inference rules. Fig. 10. Calculation of RPN by fuzzy logic model
Gallab et al./ Procedia Computer Science 00 (2019) 000–000 8

Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235 233

Table 8. Failure scenarios of equipment maintenance with fuzzy RPN.

Equipment Failure scenarios S F D RPN

Lack of attention during a truck loading arm connection operation and opening
4 3 1 45
of the manual valve.

No disconnection of the arms before emptying and sudden breakage of the


4 1 1 25.1
truck loading arm connection.

Inadvertent disconnection due to non-change of the thread of the worn arm nut. 4 2 1 26.3
Arm
Incorrect tightening and connection of the arm connection. 3 1 1 7.79
(liquid/ gas)
Failure to carry out the annual inspection of the thread wear of the arm nut. 5 2 1 26.3

Non-closing of the arm valves during maintenance. 4 1 2 45

Lack of fat in the rubbing parts. 1 2 3 25.1

Use of an unsuitable or salty lubricant. 2 1 3 25.1

Changing the pump seal with an inadequate electrical seal without checking on
2 1 2 9.38
a pump.

Valve blockage ceases recirculation of the fluid. The cessation of the


recirculation of fluid no longer allows the maintenance of minimum flow, 4 1 2 45
which causes degradation of the pump.

Troubleshooting the pump with a leaky liner. 3 2 2 26.3

Poor alignment of the tubing causing premature wear of the bearings or


2 2 1 7.93
Pump bearings of the pump.

Abusive lubrication causing rapid heating of the bearings. 1 2 2 7.93

Forgetting to degas the trim during its repair. 2 3 2 26.4

Cutting of the packing liner O-ring when refitting the packing due to non-
3 1 2 25.1
protection of the shaft thread.

No installation of a purge point to reduce the risk of gas in the body of the
1 2 1 7.93
pump.

Start-up of a gas compressor with the isolation valves closed which leads to the
degradation of the compressor (Forgetting opening of the isolation valves when 5 2 3 55.1
starting the compressor).

Poor motor bearing due to failure to follow the manufacturer's instructions


3 2 2 26.3
during lubrication.
Compressor
Poor cleaning or non-replacement of a worn / damaged valve. 3 2 1 9.29

Absence of technical device upstream of the isolation valves. 3 1 2 25.1

Absence of traceability of the compressor consignment. 2 3 1 7.93


234 Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235
Gallab et al./ Procedia Computer Science 00 (2018) 000–000

Note that the fuzzy logic model calculates the different failure scenarios and gives a different result for scenarios
that have the same result with the normal RPN calculation (see Table 9). From these results, we notice that from risk
level calculation R gives the same result whereas the model by fuzzy logic gives values that are more meaningful.

Table 9. The comparison between R and fuzzy RPN.

Failure scenarios S F D R RPN

Poor cleaning or non-replacement of a


3 2 1 6 9.29
worn / damaged valve.

Absence of technical device upstream of


3 1 2 6 25.1
the isolation valves.

Absence of traceability of the compressor


2 3 1 6 7.93
consignment.

4. Conclusion

The aim of this work is to quantify risks and failures that may appear when performing maintenance activities in
the LPG supply chain using fuzzy sets theory. A fuzzy logic approach was adopted to handle the uncertainty
conditions.
The proposed model for quantifying the maintenance failure risk is considered as a new approach compared to
the conventional approach. The obtained results confirm its applicability of the suggested model.
The advantage of the fuzzy logic approach lies in the possibility to operate imprecise and insufficient data. It is
the most suitable theory to describe qualitatively the linguistic variables and to avoid work with numerical values.
The fuzzy logic theory incorporates linguistic variables as input values and returns a result that can be wholly
defined linguistic variable.
This approach has some limitations like its inability to define the interactions and causal links between different
failures and to detect the emergence of new risks that may influence the LPG supply chain.

Acknowledgement

The authors would like to thank SMARTiLAB laboratory and the Honoris EMSI High School of engineering group.

References

[1] A.R., Hale, B.H.J., Heming, K., Smit, F.G.Th., Rodenburg, N.D., Van Leeuwen (1998) “Evaluating safety in the management of
maintenance activities in the chemical process industry”. Safety Science, 28, 1, pp. 21-44.
[2] R.G, Baston, P.S., Ray, Q., Wan, W.H., Weems (1999) “How preventive maintenance impacts Gatlinburg TH”. Maintenance and
Reliability Centre, University of Tenessee.
[3] S., Mason (1990) “Improving plant and machinery maintainability”. Applied Ergonomics, 21, pp. 15-24.
[4] T. S., Ray, C., in Langton, C., Taylor, J. D., Farmer, S., Rasmussen (1991) “An approach to the synthesis of life: Artificial Life II”, volume
XI of Santa Fe Institute Studies in the Sciences of Complexity, pp. 371–408 (Addison-Wesley, Redwood City, CA.
[5] C., Grusenmyer (2000) “Maintenance Operation and safety interactions. Bibliographic study: Maintenance tasks: definition and
characteristics contributing to their criticality”. Scientific and Technical notes INRS, 188, 46 p.
[6] C., Grusenmyer (2002) « Interactions maintenance, exploitation et sécurité : Etude exploratoire. Cahiers de notes documentaires ».
Notebooks of INRS, I86, pp, 53-66.
[7] C., Grusenmyer (2005a) « Les accidents de travail lies à la maintenance, importance et caractérisation ». Cahiers des notes documentaires,
4ème trim 2005, INRS, Paris.
[8] C., Grusenmyer (2005b) « Les accidents de travail liés à la maintenance, Etude bibliographique, les notes scientifiques et techniques de
l’INRS, 248, Vandoeuvre, 68p, Paris.
Maryam Gallab et al. / Procedia Computer Science 148 (2019) 226–235 235
Gallab et al./ Procedia Computer Science 00 (2019) 000–000 10

[9] Giraud L., Ait-Kadi, D (2001) “Maintenance: state of knowledge and exploratory study. Activities quotes of IRSST”, Montreal (Quebec),
27.
[10] R., Sanders (2005) “Maintenance-induced accidents and process piping problems”. In Chemical process safety: Learning from case
histories (3rd ed.). (pp. 91e123) Elsevier Inc.
[11] Lind, S. (2008) “Types and Sources of fatal and severe non-fatal accidents in industrial maintenance”. International Journal of Industrial
Ergonomics, 38, pp. 927-933.
[12] K., Øien, P., Schjølberg, O., Meland, S., Leto, H., Spilde, (2010) “Correct maintenance prevents major accidents. MaintWorld, 26 and 28.
[13] J., Vinnem, R., Bye, B., Gran, T., Kongsvik, O., Nyheim, E. (2012) “Risk modelling of maintenance work on major process equipment on
offshore petroleum installations”. Journal of Loss Prevention in the Process Industries, 25(2), 274e292.
[14] J. E., Vinnem (202) “On the analysis of hydrocarbon leaks in the Norwegian offshore industry”. Journal of Loss Prevention in the Process
Industries, 25(4), 709e717.
[15] B., Gran, R., Bye, O., Nyheim, E., Okstad, J., Seljelid, S., Sklet (2012) “Evaluation of the risk OMT model for maintenance work on major
offshore process equipment”. Journal of Loss Prevention in the Process Industries, 25(3), 582e593.
[16] P., Okoh, S., Haugen (2013) “The influence of Maintenance on some selected major accidents”. CEt Chemical Engineering Transactions,
Vol. 31.
[17] P., Okoh, S., Haugen (2014) “A study of maintenance-related major accident cases in the 21st century. Process Safety and Environmental
Protection, http://dx.doi.org/10.1016/j.psep.2014.03.001.
[18] J. E., Vinnem (2013) “On the development of failure models for hydrocarbon leaks during maintenance work in process plants on offshore
petroleum installations”. Reliability Engineering and System Safety, 113, 112e121.
[19] L., Barbet, J.C., Blaise, C., Bonnaud, J.P., Caillet, C., Grusenmyer, J.L., Pomian, O., Tierno (2013) Maintenance of risk activities ED123.
Sheets INRS Safety Practices.
[20] M., Gallab, H., Bouloiz, M., Tkiouat (2017) “Towards a model for developing an information system as a decision support to risk
assessment”. International Journal of Industrial and Systems Engineering, Vol. 25, N°1, pp. 110-129.
[21] M., Gallab, H., Bouloiz, E., Garbolino, M., Tkiouat, M., ElKilani, N. Bureau (2017) “Risk analysis of maintenance activities in a LPG
supply chain with Multi-Agent approach”. Journal of Loss Prevention in the Process Industries, Vol. 47, May 2017, pp. 41-56.
[22] K., Shang, Z., Hossen (2013) “Applying fuzzy logic to risk assessment and decision-making”. Canadian institute of actuaries.
[23] F., Dernoncourt (2013) “Introduction to fuzzy logic”. Massachusetts Institute of Technology.
[24] F. Dernoncourt (2011) “Introduction to Fuzzy Logic”, p.14.
[25] L. A. Zadeh (1965) “Fuzzy Sets,” Information and Control, Vol. 8, No. 3, pp. 338-353. http://dx.doi.org/10.1016/S0019-9958 (65)90241-X.
[26] L. A. Zadeh (1975) “The Concept of a Linguistic Variable and Its Application to Approximate Reasoning,” Information Science, 1975, pp.
199-249.
[27] L. A. Zadeh (1978) “Fuzzy Sets as a Basis for Theory of Possibility, International Journal Fuzzy Sets and Systems, Vol. 100, No. 1, 1978,
pp. 3-28, 1978.
[28] A., Grassi, R., Gamberini, C., Mora, B., Rimini (2009) “A fuzzy multi-attribute model for risk evaluation in workplaces”. Safety Science,
47, 707–716.
[29] P., Zalewski (2011) “Risk assessment of LNG carrier systems failure using fuzzy logic”. Scientific Journals Maritime University of
Szczecin, 25(97), 77–85.
[30] Z., Zhang, X., Chu (2011) “Risk prioritization in failure mode and effects analysis under uncertainty”. Expert System with Applications, 38,
206–214.
[31] D., Petrovic, M., Tanasijevic´, V., Milic´, N., Lilic, S., Stojadinovic´, I., Svrkota (2014) “Risk assessment model of mining equipment
failure based on fuzzy logic”. Expert Systems with Applications, 41 (2014) 8157–8164, 2014.
[32] R.M., Chandima Ratnayak, K., Antosz, “Development of a Risk Matrix and Extending the Risk-based Maintenance Analysis with Fuzzy
Logic”. Procedia Engineering 182 (2017) 602 – 610.
[33] N., Sabina, J., Zarifa, N., Azita (2012) Using Fuzzy Decision Support Systems in Human Resource Management. International Conference
on Innovation and Information Management, IPCSIT, vol.36.
[34] M., Gallab, H., Bouloiz, M., Tkiouat, E., Garbolino, M.A., ElKilni (2016) “Simulation of Failure Scenarios related to Maintenance
Activities”. International Journal of Applied Engineering Research, Vol: 11, N°24, 2016, pp. 11935-11949.
[35] M., Gallab (2017) “Développement d’une approche d’aide à la maitrise des risques dans les activités de maintenance d'une chaine logistique
: Approche par modélisation et simulation basée sur les systèmes multi-agents”. Gestion et management. PSL Research University.
Français. 〈NNT : 2017PSLEM028〉. 〈tel-01745895〉.
[36] Y., Lamrani Alaoui, M., Tkiouat (2017) “Managing Operational risk related to microfinance lending process using fuzzy inference system
based on FMEA method: Moroccan case study”. Scientific Anals of Economics and Business. 64(4), 459-471.

You might also like