Professional Documents
Culture Documents
Ways To Break Out: of The Reactive Maintenance Cycle of Doom
Ways To Break Out: of The Reactive Maintenance Cycle of Doom
com
5 WAYS TO
BREAK OUT
of the Reactive
Maintenance
Cycle of Doom PAGE 14
BREATHING NEW LIFE INTO OLD VALVES PAGE 18 SMART ONLINE MONITORING PAGE 28 OVERVIEW OF DUTCH WORKING CONDITIONS PAGE 48
While were known for our world-class motion control products, Moogs ability to
op there. Moog Industrial Services provides
maximize your machine uptime doesnt stop
hat match your specific O&M needs and
products, services and total support that
machinery challenges.
y a Moog maintenance
Download Lessons Learned by
expert at info.moog.com/mw.
REPLACEMENT PRODUCTS/SPARES
Ensure like-new performance
with new, authentic Moog
replacement products that meet
todays technical specications.
FACTORY REPAIRS
Rely on high-quality servo repair
me.
for less unplanned downtime. LOCAL SUPPORT
Work with responsive, locally-
based Moog teams who speak your
language and know your industry
and
d equipment.
i
HANDS-ON TRAINING
Gain motion control expertise
and effectively manage installation
and troubleshooting.
moogglobalsupport.com
EDITORIAL
Digitalize or die
RECENTLY, I WATCHED an interesting keynote presentation at a confer-
ence with a topic Uber yourself before you get Kodaked. This keynote
was not presented at a B2C conference or in a future-oriented manage-
ment forum. Instead, the theme of losing your business if neglecting
the digitalization around you was discussed in PaperCon, the worlds
largest technical conference for the paper and packaging industry.
Even the more conservative industries need to be aware of the fact
that digitalization can and will transform the ways to do business. It
might be that the core business stays the same, but in order to be more
efficient, to generate new revenue streams, to integrate yourself more
deeply into the value chain of your customer, and
most importantly, to gain or retain competitive
advantage, you need up-to-date information.
Information about what has happened, what is
Information about what has happened,
happening, and what will happen are at the core of what is happening, and what will happen
digitalization. Knowledge, indeed, is power. Power
to direct your business into a more sustainable
are at the core of digitalization. Knowledge,
future with competitive advantage.
However, activities related to digitalization are
indeed, is power.
not valuable per se. Every action and choice you
make should help you build a better business and better operations
than before, just like in all other investments. Keeping this in mind,
it is important to know that many digitalization-related ideas can be
tested and piloted quickly and at a fraction of the cost of machine in-
vestments. Consequently, even big companies can now validate their
ideas quickly, or alternatively fail fast.
Once you know through pilots, what areas of digitalization work for
you, it is easier to make the final decision to invest in integrating the
areas into your routine operations. Come to think of it, five years from
now, no one will be talking about digitalization anymore; it will be busi-
ness as usual.
Emil Ackerman
Managing Director & Co-Founder of Quva Oy, Chairman of the IIoT
Committee of the Finnish Maintenance Society Promaint
14
Planned and scheduled
work is 20 percent
more efficient than
unmanaged work. And
it is safer.
4 maintworld 2/2017
IN THIS ISSUE 2/2017
10
In mountainous tunnels under
high rock cover, the true ground
conditions can be difficult to
predict, even with information
collected from an exploratory
tunnel.
38
Mining operators are
increasingly confronted
with security breaches
on their properties.
Issued by Promaint (Finnish Maintenance Society), Messuaukio 1, 00520 Helsinki, Finland tel. +358 29 007 4570 Publisher
Omnipress Oy, Mkelnkatu 56, 00510 Helsinki, tel. +358 20 6100, toimitus@omnipress.fi, www.omnipress.fi Editor-in-chief
Nina Garlo-Melkas tel. +358 50 36 46 491, nina.garlo@omnipress.fi, Advertisements Kai Portman, Sales Director, tel. +358 358
44 763 2573, ads@maintworld.com Subscriptions and Change of Address members toimisto@kunnossapito.fi, non-members
tilaajapalvelu@media.fi Printed by Painotalo Plus Digital Oy, www.ppd.fi Frequency 4 issues per year, ISSN L 1798-7024, ISSN
1798-7024 (print), ISSN 1799-8670 (online).
2/2017 maintworld 5
INDUSTRIAL INTERNET
XXXXXX
Combining Machine
Learning with IIoT Should
Be Your Priority
Machine learning and IIoT are no longer nice things to have, but should be applied
together to reap the rewards of cost savings, improved uptime, and performance.
In this first of a two-part series on machine learning, Richard Irwin, senior product
marketer with Bentley Systems, explains what benefits machine learning can bring
to asset-intensive industries, in relation to the Industrial Internet of Things.
THE INDUSTRIAL WORLD is awash with movie services that recommend titles to
data and new information from sensors, watch based on viewing habits, to banks
RICHARD IRWIN, applications, equipment, and people. that monitor spending patterns to detect
Bentley Systems,
richard.Irwin@bentley.com
But the data is worthless if it is left un- fraudulent activity. Now, the industrial
touched or not used to its full potential arena is moving quickly toward using
with the latest technology. To make the machine learning to take advantage of
most of big data, industry leaders should the Industrial Internet of Things.
implement machine learning along- As the velocity and variety of data
side the Industrial Internet of Things becomes available through advance-
(IIoT) to take advantage of the benefits ments in sensor technology to monitor
increased information can bring to any just about anything, machine learning
organization that is asset and data-rich. is being applied to efficiently manage
We have all experienced some form increasingly large and fast-moving data
of machine learning, from streaming sets. Machine learning can handle large
6 maintworld 2/2017
INDUSTRIAL INTERNET
and complex information to discover data that is out of sync, like a piece of
otherwise hard-to-see patterns or trends equipment that is underperforming and
in the data, such as normal behaviour could potentially fail. It is then able to
and anomalies. It can then learn these add this piece of information and learn
patterns and apply it to new data to de- from it in future models. This makes de-
tect similar patterns in the future. An cision-making easier and more reliable.
example would be to model the perfor- In any industry, the ability to rec-
mance of a piece of equipment, such as ognize equipment failure, and avoid
2/2017 maintworld 7
INDUSTRIAL INTERNET
THE ABILITY TO RECOGNIZE EQUIPMENT FAILURE, AND AVOID UNPLANNED DOWNTIME, RE-
PAIR COSTS, AND POTENTIAL ENVIRONMENTAL DAMAGE, IS CRITICAL TO SUCCESS. THIS IS
EVEN MORE RELEVANT IN TODAYS TURBULENT TIMES.
casting techniques, utilities can ascer- right permits) and route optimization through large amounts of data, identify
tain hourly demand and peak hours for a to reduce travel and costs. These are still patterns and trends within it, and make
day, allowing them to optimize the power done manually in some cases, and are predictions. Key decision makers need
generation process. Using information error-prone. Machine learning can do to take advantage of this match made in
such as historical demand data, regions, this automatically, and by incorporating heaven to optimally control and man-
population, weather patterns, events, historical data of previous jobs, weather, age their assets get on top of it now
and so on, organizations can predict de- and time of year, it could gauge where before asset data is unmanageable. Pre-
mand on any day or period in the future. problems might arise. venting assets from failing, forecasting
This is essential for ensuring the utility demand, and optimizing movements of
can produce the resource in time. Summary your workforce, are just a few of the ma-
Workforce Management In With the arrival of the Industrial Inter- chine learning techniques that can be ap-
workforce management, the aim is to net of Things, data is growing and be- plied to improve the overall performance
optimize the workforce and make it coming more accessible. With the ability of the organization. By combining these
as productive as possible while reduc- to acquire more data, more advanced elements and more, machine learning
ing labour costs, travel, additional call technologies are required to scrutinize will bring, as it matures, significant ben-
outs and improving customer service. and filter out the important information efits. The Industrial Internet of Things
There are measures in place to help with and value held within. But, it can only and machine learning should no longer
smarter scheduling (mapping the right be exploited by identifying what works be considered just buzzwords instead,
person with the right skills to the right well and what does not. Machine learn- put together, they should be your num-
location with the right tools and the ing features complex algorithms to sort ber one priority.
8 maintworld 2/2017
TECHNOLOGY
10 maintworld 2/2017
TECHNOLOGY
cutterhead. Open-type machines are provides protection from rock falls and assistance of the ring beam erector or
capable of operating in ground with oc- other problems, making it ideal in hard, some other mechanical means. The most
casional to continuous water as long as a blocky ground as well. In addition, in desirable location to place such support
mitigation strategy combining grouting squeezing ground Double Shield TBMs is immediately behind the cutterheada
to stem flows, as well as pumps to re- can be used with compressible material problematic situation with a shield type
move the water, is employed. as backfill or special segments to accom- machine. The machines also must be
modate squeezing conditions. equipped with very high torque to over-
Shielded Hard Rock TBMs Both types of machines can be suc- come the squeezing effect.
Most shielded TBMs line the tunnel cessfully utilized in a wide range of con- If using a shielded machine erecting
either simultaneously with or directly ditionseven in squeezing ground and segments, certain features such as a con-
after a TBM stroke, resulting in an ear- significant water inflowsif properly vergence measuring systema hydraulic
lier useable date for the tunnel. Shielded designed. A host of technology, termed cylinder mounted on top of the machine
machines also have the very beneficial Difficult Ground Solutions (DGS), can and connected to the machines com-
advantage of providing a limited section be used on these machines, from multi- puter system (PLC)can detect when
of non-heavy support; i.e., the distance speed gearboxes that enable excavation squeezing conditions are present. Hav-
from the cutterhead to the grouted lin- in fault zones to shield lubrication and ing a machine designed with the shortest
ing. Shielded TBMs can also have diffi- breakout thrust/torque for squeezing possible shield length, and a stepped
culty in faulted rock, as the working area and collapsing ground. (tapered) shield if necessary can be im-
for ground consolidation can somewhat mensely helpful. As mentioned previ-
restrict good face coverage. There are ADVANCEMENTS IN ously, shield lubrication and emergency
two types of shielded hard rock TBMs: thrust can get a machine through a situ-
Single Shield and Double Shield. GROUND SUPPORT ation where it might otherwise become
Single Shield TBMs are shorter in trapped.
length and can therefore be launched Squeezing/Convergent Ground
from a shorter starter tunnel, and are For squeezing or converging ground, Rock Bursting
typically utilized in non-self-supporting over-boring is often necessary. The In rock bursting conditions wire mesh
rock, as the machine advances by react- only practical solution to over boring with rock bolts, yielding rock anchors,
ing against the concrete tunnel lining is to pre-mount extra gage housing in steel arches, ring beams or combinations
rather than unstable tunnel walls. They the periphery of the cutterhead. In the of all the above may be required. Such
have the disadvantage of not having grip- over-bore zone, yielding type structures support can be placed with rock drills,
pers, which allow greater pull, thrust and should be erected if using an open-type a ring beam erector, and a shotcrete
jogging of the cutterhead. machine. These structures can include system. It is important to hold the rock
Double Shield TBMs are ideal in self- yielding steel arches, steel arches in con- in place to control and limit the distur-
supporting rock, and some non-self-sup- junction with yielding jacks, shotcrete bance of the rock to as great an extent
porting rock, or in combination ground structures with yielding rock anchors, as possible. Rock bursting could also be
since they can react against either tun- or combinations of the above supports. contained with TBMs in association with
nel walls or segments. The shield also Such support needs to be placed with special lining.
2/2017 maintworld 11
TECHNOLOGY
In squeezing or blocky ground, shield lubrication allows a When significant water inflows occur, a guillotine gate can seal
shielded rock machine to get through geology where it might off a shielded machine to allow crews to safely control the
otherwise become stuck. water (sealed area in blue).
With modern open-type TBMs, Fault Zones & Water Pressure significant water inflows). However, if
ground support such as the McNally Fault zones can be the most difficult there is a possibility of significant pres-
Roof Support System can be used to condition to encounter, especially when sures and/or a massive inrush of water,
allow lining to be extruded from the ma- associated with water under pressure. then a shielded machine with DGS fea-
chine as it advancesa very safe option They are also the most difficult condi- tures is recommended. In the event of
in these conditions. Todays TBMs are tions for predicting expected advance a large inrush of water, a guillotine gate
also equipped with all of the same tools rates. on the muck chute can effectively seal
and techniques that are used in drill & In all conditions, advance probe drill- off the muck chamber to keep the crew
blast operations to excavate through ing is recommended 30 to 40 meters in safe as well as keep the machine from
difficult rock conditions. With sophis- advance of the face with a 10 m overlay. becoming flooded out. Additional inflat-
ticated probing techniques installed on This is especially important when fault able seals can seal the gap between the
the TBM, the operator can predict what zones or water are expected. When a telescopic shield and outer shields of a
is ahead of the tunnelling operation fault zone or water is encountered, the Double Shield TBM to keep everything
more quickly than drill & blast and react extent of the zone should be explored watertight. This system is termed pas-
appropriately. On a shielded machine, prior to TBM boring within 10 20 me- sive water protection because the TBM
probe drilling is equally important. The ters of the zone. Drilling should be done is stopped in place (not actively oper-
machines shield provides safety against on a 360-degree basis. First, the zone ating). During that time the crew can
rock bursting events. should be grouted to stop water inflows. then work to grout off water inflows and
After grouting, ground consolidation dewater the chamber to control the flow
Swelling/Slacking Ground additives should be injected into the before they begin boring again.
In swelling and slacking conditions unstable rock or soil material. It may
an effective ground treatment is shot- be necessary to inject such material Blocky or Jointed Rock
crete applied immediately behind the into the face at short intervals of 2 to 4 In blocky or highly jointed rock, the
cutterhead, for both open-type and meters, and advance at shorter intervals. McNally system to hold the rock in place
shielded machines. In extreme condi- The support of geologists experienced in has been proven very effective in open-
tions, over-boring may be required and predicting and treating fault zones, and type machines. If the rocks are held in
measures for rock support in squeezing of ground conditioning experts, is highly place then this can prevent or lessen the
ground may be needed. The support recommended when fault zones are en- condition of cathedralling over the cut-
can be a combination of shotcrete, rock countered. terhead and fallout in front of the face;
drills and ring beams on an open-type For passing through fault zones, grout it will also reduce cutterhead damage.
machine. The difficult question, how- and ground conditioning holes are re- The ground support should be placed
ever, is to predict the extent of swelling quired. After ground treatment, ground as close as possible to the cutterhead.
and squeezing. This is a very important support such as spiling or forepoling Rock supports for the McNally system
consideration when considering the through the front shield over the cut- can be prefabricated rebar, wood/metal
use of concrete segments in such condi- terhead may be necessary for safe and slats, or wire mesh in conjunction with
tions. Because of the difficulty of pre- predictable advance. It is preferable to rock straps and rock bolts. In a shielded
dicting the extent of swelling, two-pass carry on this drilling as close to the face machine, DGS features previously
lining systems have been used such as as possible to ensure good face coverage. mentioned including shield lubrication,
in the large diameter Niagara Tunnel These methods are all possible whether tapered shields, and hydraulic shield
Project in sedimentary rock. This large an open-type or shielded machine is breakoutwhere radial ports in the
diameter (14.4 m) tunnel utilized initial used. machine shield can be made to inject
ground support followed by a slipform When water is present in a hard rock pressurized hydraulic lubricants to free
concrete liner and a waterproof mem- tunnel, it can be pumped away from the a shield that has already become stuck
brane. face and out of the tunnel (even fairly are all useful.
12 maintworld 2/2017
CONDITION MONITORING
5
Ways to Break Out of
the Reactive Maintenance
Cycle of Doom
Improving reliability will help the bottom line of any organization and much
more. However, if that organization is experiencing any level of reactive maintenance,
then improved reliability may seem like a pipe dream. Until you can break out of the
reactive maintenance cycle of doom, it is not possible to make any real inroads into
your reliability improvement initiative.
14 maintworld 2/2017
CONDITION MONITORING
And what is managements response? cycle so that we can make real improve- analysis process. And then we could
Do they reduce the headcount and try to ments in reliability? implement that plan and ensure that we
squeeze the budget in order to deal with do what is necessary to change peoples
the high costs of downtime and reac- 1. First We Have to behaviour.
tive maintenance? What happens then? Change the Culture But people do not like to be changed.
Does morale decline? Do standards drop The first step is to change peoples at- They will change when they are a con-
further? And as a result do you experi- titudes, which will change peoples tributor to change.
ence even more preventable failures behaviour. Everyone has to believe that Besides, who knows more about the
that consume resources and results in they will be better off in a reliable plant. reasons why equipment fails (and why
temporary repairs being performed? We They also have to believe that they can we experience production slowdowns)
could continue to go through this list, contribute to the reliability improve- than the operators of the equipment
around and around and around. ment process. and the front-line people who maintain
Well, if you answer yes to the major- There are a number of ways to go it? But do we normally ask those people
ity of the questions above, then you are about doing this, and it is a topic worthy for their opinion or assistance? No, we
trapped in the reactive maintenance cy- of its own article (or even its own book) dont And then we wonder why we
cle of doom You have to get out! but there are a few things we can do to dont enjoy the success we hope for.
improve the culture: So that is something that we are go-
But How Do You Get Out? 1. Ensure that senior management ing to do differently in the future. The
We have to get out of the reactive main- is 100 percent behind reliability brown-paper process will gather to-
tenance cycle of doom. If everyone is improvement and ensure that gether the front line people in small
dragged back to perform reactive work, they are vocal in their support groups and learn from them what needs
then none of the proactive tasks that will 2. Educate people so that they truly to change. And then we will ask them to
ultimately lead to improved reliability understand the benefits of work- lead the mini-projects that correct the
can ever be performed. Well, you may ing in a reliable plant problems that are identified in that
believe that management could hire ad- 3. Educate people so that they un- way they take ownership and free up
ditional people so that you have the re- derstand why failure occurs your time. An activity like this is key to
sources to perform those proactive tasks. 4. Involve people, ideally in a your success.
That is about as likely as management brown paper process, in or-
hiring unicorns to maintain the grass. der to get their suggestions for 2. Implement a Work
The basic answer is that we have to do improvements and get their as- Management Programme
much more with the people and budget sistance in making the improve- Well, we really just need basic (but effec-
that we have right now. We have to stop ments tive) planning and scheduling for now.
performing tasks that waste our money, That last point needs a brief explanation. Wait a minute. Doesnt that mean you
waste our time, and induce failures in As intelligent managers or mainte- need an extra person? No, you will take
our equipment. This may sound like a nance/reliability engineers, we can come one of your most effective trades people
push for higher productivity. In a sense, up with a lot of ways to improve reli- and put them in the role of planner/
it is. But that is not our focus. ability. We can devise a strategy via the scheduler. But surely that must mean that
So, how do we break out of this vicious Reliability Centred Maintenance (RCM) you have even fewer people to perform the
corrective (and hopefully proactive) main-
tenance work. That is true, but the fact is
that we will make the remaining trades
CONDITION MONITORING WARNS YOU ABOUT PENDING people far more effective.
EQUIPMENT FAILURES; THE NATURE OF THE PROBLEM AND Planned and scheduled work is 20
percent more efficient than unman-
THE SEVERITY OF THE PROBLEM. aged work. And it is safer. And the work
will be done right the first time and
Follow a Roadmap to
Tackle the Reactive
Maintenance Doom
If you do not have a strategy,
then feel free to follow the
chart below. It has been
developed to ensure that
you lay a solid foundation
before you tackle the reactive
maintenance cycle doom,
and then you can focus on
the worlds best practice of
reliability improvement and
operational excellence.
2/2017 maintworld
i ld 15
CONDITION MONITORING
3. Work on Communication
to Foster Cooperation
The chances are that the relationship be-
tween maintenance and operations/pro-
duction is not great That relationship
has to improve. The maintenance team
needs the cooperation of operations so
that the equipment is ready when work
must be performed. Operations needs
to work closely with the maintenance
team to ensure that their equipment
can perform with minimal downtime,
minimal slowdown, and the highest level
of quality.
In addition, the operators of the
equipment can contribute to reliability
improvement by changing the way they
operate the equipment, and performing
basic inspections and maintenance tasks
16 maintworld 2/2017
Be a LUBExpert
h^ sdtultrasound.com/lubexpert
XXXXXX
CONDITION MONITORING
18 maintworld 2/2017
CONDITION
XXXXXXXXXXXXXX
MONITORING
will turn to Moog instead of an unau- pressure and the increased temperature ness by asking a potential customer to
thorized repair house (URH). As part as the hot oil was prematurely returned compare the repair price per valve with
of the program, even if a maintenance to the sump. that of an OEM, the maintenance cost
manager has a Moog valve that turns out The mill returned four servo valves added into a URHs price ultimately
to be obsolete, Moog will clean the valve, for repair. Moog determined that the makes it more economical to have the
evaluate its condition and provide users internal leakage was eight times greater OEM service your servo valve. It is true
with an inspection report describing the than it recommends. After inspection, that you can go to a URH and get what
health of the valve as well as recommen- technicians learned that some of the they would state is a serviced valve. We
dations for potentially replacing it. spools were not made by the Moog and have seen time and again though, that
Turning to an OEM to affect repairs were not manufactured to the correct these URH-repaired valves typically last
or service equipment is a logical step. tolerances. These were also an older between six and twelve months until a
However, some maintenance manag- style valve with a mechanical null ad- plant manager will see an unstable pro-
ers do just the opposite because they justment. Moogs team upgraded the duction platform caused by early fatigue
believe asking the OEM to repair a piece valves to a more reliable magnetic null in the valve.
of equipment will break their budget. adjustment mechanism (at no additional Ultimately, that kind of repair
causes poor product quality and higher
scrap counts in production applications,
or an unplanned outage in power genera-
THE RTP PROGRAM GIVES MAINTENANCE PROFESSIONALS tion and flight simulation. Any one of
ACROSS A VARIETY OF INDUSTRIES THE SERVICES TO these things could lead to thousands (or
hundreds of thousands) of dollars in lost
EVALUATE AND REPAIR THE SERVO VALVES RUNNING THEIR product or productivity.
In contrast, when maintenance man-
EQUIPMENT agers turn to an OEM to have their valves
repaired, Moog, in particular, carries out
these repairs by following ISO stand-
The viewpoint of some maintenance charge). The valves were installed on the ards. Fastidiously adhering to industry
professionals is that if they turn to an gauge line and the pressures returned to standards, in turn, ensures that a plant
unauthorized repair house, or even by- normal with only three pumps operat- or equipment operator can count on his
pass an expensive repair altogether, they ing. Temperatures returned to normal or her repaired valve running for a bil-
will save money. Our experience is just too. This saved the company almost lion cycles, which correlates to years of
the opposite. US$50,000 a year in electricity. And the service.
For example, a steel mill had difficulty valves have now been in service for more The selection of high quality com-
maintaining hydraulic system pressure than two years without the mill having ponents is critical to minimizing un-
on its gauge line. Its HPU system was to return them to Moog for repair, which planned downtime. The cost of opera-
robust with three 33-percent capacity has yielded significant savings in main- tional loss during unplanned downtime
pumps and a fourth pump in reserve. tenance costs. whether it is industrial production or
As the situation worsened, the mill was transportation justifies the time it takes
forced to operate the reserve pump, con- Think Twice When an engineer to evaluate the life expec-
sume additional electricity and operate Considering a URH tancy of critical components on equip-
without a spare pump. At the same time, So while a URH might try to win busi- ment. For example, test stands and flight
the pressures were declining and the
fluid temperatures were increasing. The
mills managers approached Moog and
its local distributor for assistance. Moog
audited the system and learned that the
servo valves had not received a Moog
factory repair in many years. Instead, a
URH had repaired the valves.
Moog believed that the root cause
of the steel mills problem was leakage
through the servo valves caused by worn
seals between the bushing and the spool. A Moog technician
installs a new
Moog maintains strict leakage standards
electronics board
for repairs and will replace the bushing on a servo-
spool assembly if the leakage is exces- proportional valve.
sive. A URH cannot purchase replace-
ment parts from Moog and will reuse
worn parts as long as the valve is func-
tional. The severe leakage explained the
higher flow needed to maintain system
2/2017 maintworld 19
XXXXXX
CONDITION MONITORING
simulators are valued in the millions of cant. URHs also lack the documentation mend an equivalent Moog valve with
dollars and the unexpected failure of a to test a repaired part against the origi- aggressive trade-in pricing that meets or
servo valve or actuator translates into nal specifications, which is a hallmark exceeds any competitors specifications.
lost revenues and schedule delays that of quality repair. As part of Moogs RTP For plants and factories that might
can often exceed the original equip- Program, if a maintenance manager have a large base of installed valves
ment costs. Early fatigue in the internal sends us a valve for repair and we note spanning many makes and models,
servo valve components generally do not that our engineering team has upgraded Moogs engineers can identify the proper
exhibit any external signs of degrada- that particular model of servo valve, replacements and develop a plan to
tion such as leakage. However, the early we will upgrade the valve sent to us for minimize the number of valve models
fatigue definitely affects the dynamic repair as part of our standard services. in operation at a facility. We like to refer
performance of the servo valve. Once Practically speaking that might mean to this analysis and planning as a way to
the servo valve is mounted in the equip- our repair group would receive an older reduce valve sprawl. By reducing vari-
ment, most equipment operators will valve with a steel ball on the feedback ous makes and models of valves inside a
not be able to visually see any perfor- wire, matched with a slotted spool. We plant and relying instead on a common
mance changes unless the system fails would take it upon ourselves to upgrade model for spares, maintenance managers
catastrophically. the valve to incorporate a carbide ball on can reduce their inventory rate.
For plant managers who still opt for the feedback wire matched with the ball- Truly repairing your valves requires
an unauthorized repair house to solve in-hole spool design. Enhancements like inspecting the hardware and providing
their problems, they might be surprised this increase the longevity of the valve. an inspection (or repair) report that
to learn that sometimes the equipment identifies what is wrong with the valve
in question still ends up in the OEMs Calling All Valves and how the valve is functioning. With
hands. We have seen instances in which If a plant manager has valves in their our experience analyzing and repairing
a URH couldnt repair a Moog servo equipment that are very old or obso- valves, we can also solve a plant man-
valve, and they sent it to us in pieces. We lete, the RTP program is still an option. agers problems on a system level, not
serviced the valve and returned it to the Moogs repair technicians and engineers just the component level. This is some-
URH. If the plant or mill operator had will evaluate these older makes and mod- thing else a URH is not qualified to do. If,
come directly to Moog or used a Moog els of valves to assess the performance for example, we see valves come back to
authorized distributor, then we would against the original specifications. Some us with worn spools caused by repeated
have saved them time and money. of the assembly and test technicians who contamination or a cracked flexure
In fact, when time is critical, we have conduct this work for Moogs custom- sleeve, our technicians can identify what
actually repaired a Moog servo valve and ers have been in their roles for up to 30 is wrong with the system and recom-
returned it by plane to its owner on the years, so they have seen everything. Once mend a system-wide solution.
same day we received it. And while speed they analyze a valve, the technicians and And that is a guaranteed way of get-
is sometimes of great importance, the engineers can recommend if the valve ting the most from your valves and
quality of a repair is even more signifi- can operate safely and, if not, recom- boosting productivity.
20 maintworld 2/2017
1/2017
Master the Language
of Your Machinery
Optional modes:
Adash
#|r
=qq
}J|}
t
Your local distributor can
q
be found on our website
info@adash.com www.adash.com under
www.adash.com Contacts.
XXXXXX
RELIABILITY
22 maintworld 2/2017
Power Up Your Skills
Open Courses in Europe
Location Dates Language The key element to any operation leaders success is having a maintenance
NL - Eindhoven 23.-24. May Dutch GHSDUWPHQWUXQQLQJDWRSWLPXPHFLHQF\7RDFKLHYHWKLVLWLVHVVHQWLDOWR
NL - Amsterdam 30.-31. May English
have competent work preparers/planners that can ensure the maintenance
DE - Dsseldorf 30.-31. May German
DE - Stuttgart 20.-22. June German
ZRUNIRUFHLVIXQFWLRQLQJHFLHQWO\HHFWLYHO\DQGIURQWOLQHOHDGHUVKLSWKDW
DE - Frankfurt 27.-29. June English schedules and executes the plan.
DK - Copenhagen 05.-07. September English
NL - Eindhoven 12.-14. September Dutch PSURYHZRUNSURFHVVHHFWLYHQHVV
DE - Frankfurt 19.-21. September German 8QGHUVWDQGWKHGLHUHQFHRISODQQLQJYVVFKHGXOLQJ
Identify leading and lagging indicators
/HDUQZD\VWRDFFHOHUDWHDQGLPSURYHWKHTXDOLW\RI\RXUSODQQLQJHRUWV
+31.30.799.5050
europe@nexusglobal.com
RELIABILITY
XXXXXX
Table 1:
Now, really think about this. What How many spare parts do you have in
component would truly shut your plant your storeroom for these assets? In fact,
down immediately? The entire opera- Id be willing to wager that the PM and
tion, the whole facility? A room full of PdM program on these assets dont re-
production people would probably come ally reflect the absolute criticality they
up with a list of production machinery; have to the operation and livelihood of
that makes sense. Maintenance and en- your facility. That, however, is another
gineering might come up with another, topic for another time.
completely different list. Very few com- The sample table shown in table 1 is
ponents send everyone home. In fact, meant as a vehicle that the group can use
its very likely that the only components to agree on the level of importance that
that would send even the receptionist the component has on the production
home are related to power distribution. of the plant. Not all components have a
Generally, failure of the first transformer weight of 5. In fact, some have no impact
you own (after the utility companys at all (-2).
transformer), or some high-voltage This addresses downtime. So, what
switchgear, can send everyone home. about costs? The same approach is
taken to consider the cost of a compo-
nent. For the sake of this article, I will You might guess, by now, that for the
Table 2: subjective thought of lead time, you are
provide another sample table. These
Cost (includes purchase Weight tables are only to be considered as sam- to construct a similar table, with break-
price and transportation ples. As I said earlier, a critical spare part out categories that make sense to your
cost) is whatever we say it is. We just have to organization. Just a side note here, if
say what it is. We do that by expressing your organization practices the Equip-
>$100,000 5
the significance through these tables, ment Design Excellence concept of parts
$50,000 - $100,000 3 standardization, you can really and posi-
and the collective sum of their weighted
$25,000 - $49,999 2 tively affect the lead time of very impor-
values. For other attributes of a critical
$10,000 - $24,999 1 spare part, simply construct additional tant components through the practice of
<$9,999 -2 tables. Strategic Spares.
For the tables reflecting loss of pro-
duction, costs, and lead time, Id suggest
five levels to consider, with correspond-
ing weights. The weights are added, with
a combined value in excess of a set point
meant to be a determinant for a critical
part. That set number is whatever your
group says it is.
But, were not done. What separates
the majority of companies, from those
truly practising world-class, or high-per-
forming, MRO practices, is the inclusion
of several more factors.
Before we venture into some addi-
tional characteristics separating normal
spare parts from critical spare parts,
consider this one very important, and al-
beit, subjective featurehow bad would
it be if you actually have to use the criti-
cal spare part?
24 maintworld 2/2017
XXXXXXXXXXXXXX
2/2017 maintworld 25
Road Map to Operational
Readiness and Asset
Performance
Ensure a Safe, Reliable, and Compliant Operation
Achieve business goals with a risk-based approach to asset management. Bentley will help get you there.
Leveraging 30 years in design and visualization innovations, Bentley is at the forefront of engineering software.
Combining 2D/3D plant models and point clouds with engineering information and asset performance management,
Bentley delivers an enterprise platform to manage assets throughout their entire lifecycle.
The visual workow supports both greeneld and browneld operations; bridging the gap between CAPEX and OPEX
and enabling a sustainable business strategy for operational excellence and safety.
2017 Bentley Systems, Incorporated. Bentley, the B Bentley logo, and AssetWise are either registered or unregistered trademarks or service marks of Bentley
Systems, Incorporated or one of its direct or indirect wholly owned subsidiaries. Other brands and product names are trademarks of their respective owners.
Assess risk based on failure severity, likelihood scores, Visually navigate assets.
and condence assessment.
Smart Online
Monitoring
Online vibration monitoring is essential
for critical machines. There are machines
that need to be monitored continuously;
it is not enough to take a measurtement
every week, or even once per day.
28 maintworld 2/2017
VIBRATION
XXXXXXXXXXXXXX
DIAGNOSTICS
Adaptive Algorithm
of Data Acquisition WE MEASURE REALLY CONTINUOUSLY AND THERE IS NO WAY
To avoid the disadvantages mentioned
above, the Adash company has devel- WE COULD MISS ANY IMPORTANT CHANGE OF VALUE
oped Adaptive Model of Data Acquisi-
tion for online monitoring. Lets go back
should be made. We usually set up the changes, more measurements would
to our first question. How often do we
significant value change to 5 percent. So be saved, but they would be important
need to measure the vibration value?
if the measured values havent changed measurements that are worth storing.
The ideal answer is continuously; it
more than 5 percent compared to the The Adash online monitoring system
means starting a new measurement
first value, there is no need to save the A3716 with new adaptive algorithm is
immediately after the previous one is
measured data, it is not important to controlled by Adash DDS software. You
finished. If the measurement takes 1 sec-
us. Why was 5 percent chosen? Because can run the database system in DDS soft-
ond, we should measure every second.
from experience we know that there is ware even on a standard laptop.
This is ideal, so lets really do it this way!
never anything really important happen-
We measure continuously and there is
no way we could miss any important
ing in this 5 percent range around the Why is it called adaptive?
measured value. See from the pictures The system can adaptively change some
change of value.
how it affects the actual saved data. of its parameters. When you measure
Ok, now you may ask what about all
the machines where the vibration values
the measured data? We have said the What Happens if Nothing
change a lot and it is common for their
database size would grow rapidly if the Changes Significantly?
run, the adaptive algorithm sets up the
measurement time were increased from When the values dont change over a re- significant value to a larger value so that
every 5 minutes to every second. Isnt it ally long time, only one measurement it does not save insignificant data. Look
insane? No, it is not! The trick is hidden would be recorded, which is also not at the picture showing how it affects
in data saving. The measurement could ideal. There is a time interval for saving the saved values. You can see that even
be taken for example every second, but data defined in the adaptive algorithm when the significant change is set at 50
why would we need to save all the meas- system, which represent the longest percent all really important values stay
ured data if the values are still the same? time I can have without data being saved.
So here comes the trick: the model of recorded.
When I set up this time interval for sav-
adaptive algorithm for data acquisition. This adaptive change of parameters
ing to e.g. 4 hours, I save a measurement
This adaptive system decides which val- can be done automatically by the system
every 4 hours even if the value has not
ues need to be saved and which values itself, but if the user wants to be cer-
been changing significantly. I believe you
dont. So lets say I take a measurement tain of the set up, it can also be entered
can see the difference between a conven-
every second for the 5 minutes and the manually.
tional system and the new adaptive algo-
value is still 3 mm/s. I have measured rithm already. Imagine how much data I have not described all the decision
this value 300 times during those 5 we would store just during one day with parameters of the A3716 adaptive algo-
minutes, but I dont need to save it 300 a conventional system; during 24 hours rithm as some of them are more compli-
times! It is enough for me to save it just it would be 288 saved measurements cated. However the algorithm does those
once and this is exactly what adaptive when taking measurement every 5 min- decisions for you and you get the results
algorithm is doing for me. utes and we still cannot be sure that we which you need. It measures really
havent missed anything. With the adap- continuously and saves just the values
How does the adaptive tive algorithm, during those 24 hours, we which are important. Some users may
algorithm select which data to would get only 6 measurements stored be little scared when they see the graph
save? if there is no significant change (when of recorded values and they see just 2
Take a closer look on how the adaptive the time interval for saving is set up to values, for example. They are afraid that
algorithm selects which data to save and 4 hours) and we would be sure that we just 2 values were measured, but no! The
which to erase. As mentioned earlier, the havent missed anything as we are re- system was measuring continuously, but
adaptive algorithm does not save the val- ally measuring continuously. The size only 2 values were worth to save. You
ues that are not changing significantly. of the database is much smaller in this can be sure that nothing has been missed
A definition then of significant change case. Obviously if there were significant and you can sleep well
2/2017 maintworld 29
COLUMN
Reliability, Maintenance
and Safety I am no longer surprised to see reliability and
TORBJRN
IDHAMMAR, maintenance improvement initiatives abandoned
President of IDCON INC
in Raleigh NC, USA, before the substantial results, which are
www.idcon.com
possible to achieve, are delivered and sustained
for years to come.
THIS PHENOMENON is recognized by many of my colleagues with have an incident rate of below 1. In 23 years this indus-
in the reliability and maintenance management profession. try as an average reduced overall safety incident rates by
I recently came across the findings illustrated in the about 87 percent. We all know that this is because of consist-
graph below from the American Society for Training and ent long term reinforcement and training.
Development. It illustrates what happens if training is not A study by University of Tennessee shows that organiza-
followed by immediate practice and reinforcement. The tions with a high level of reactive maintenance has an OSHA
findings show that 87 percent of what you learnt is lost incident rate (Incidents per 200,000 working hours) of 4.36,
within 30 days if training is not followed by practice and re- while top performers with much less reactive maintenance
inforcement. Their findings illustrate very well why so many have an OSHA incident rate of 0.11.
reliability and maintenance improvement initiatives deliv- Imagine the same focus on training, implementation
ers good results, but only about 50 percent of the improve- and reinforcement of basic maintenance practices; could
ment potential. you have reduced preventable maintenance work and down
Reliability and maintenance improvements are one of time by 80 percent? The majority of maintenance work is
the last major improvements opportunities the Industry preventable and can also be executed in half of the time so I
have left. Everyone with access to capital can buy the same know it is possible. I have seen it happen and the key to these
equipment and technology, how productive your plant is will successes has been top management long term consistent
to a very large extent depend on the reliability of your pro- leadership, support and reinforcement. And on top of better
cess and your equipment. If your equipment runs, you make maintenance productivity, higher production throughput,
product, if it does not run your employees work harder, you you would get an even better safety record.
pay more and you are not making product. So why does top
management not reinforce that even the most basic main-
tenance practices are executed better and better over long
Learning
Reference: 2015 study of over 100 companies by University of Tennessee IDCON, INC 30 days
Reliability and Maintainability Center. (UT-RMC) Reference: American Society for Trainingand Development
30 maintworld 2/2017
There are actually three
certainties in life: death,
taxes and the need for
good maintenance.
www.marshallinstitute.com
www marshallinstitute com
Marshall Institute
001 919-834-3722 ...making good companies better
ASSET MANAGEMENT
XXXXXX
1
Compressed air is used in inspect for steam leaks is ultrasound. of the valve, but can also cause prema-
so many applications. Com- Contact ultrasound is used to main- ture failure downstream especially if
pressed air leaks become huge tain the good operating condition of the the particular valve is for example on the
energy losses in some mines there are steam traps in the system by identifying suction side of a pump.
megawatts of power used to produce those failing steam traps that are not
4
compressed air. Air leaks in pneumatics removing air, CO2 and condensate from Hydraulic systems are used
and control systems however, can become the steam system. for motion and for power and
showstoppers, bringing production to a there are many valve applica-
3
stop. Using Ultrasound for listening to Valves are used in so many tions involved here too. Failure of hy-
internal air leaks or cracks on the boom applications and are virtually draulic systems is not an option and yet
of a dragline omni-present in the mining too few businesses consider any main-
industry consider how a hydraulic or tenance practice other than breakdown
2
Steam is of major importance in water system is going to operate prop- with the corresponding huge expense of
certain mining processes con- erly without the proper operation of the downtime. Ultrasound can be used on
sider the steam injection sys- valves involved. shovels for example to listen to internal
tems used in a SAG-D (Steam Assisted Process failures tracked back to in- bypassing on boom, stick and bucket
Gravity Drainage) plant for instance. correct valve operation can create a large cylinders.
Steam is injected underground to warm amount of unwanted downtime one The inspection method for cylinders
up and soften bitumen and heavier oils particular story in a coal mine comes to is quite simple: merely place a sensor on
to make them easier to extract from the mind where an internal leak on a valve the cylinder and allow it to operate in its
earth. The production of steam is there- and also on the refurbished spare in the normal fashion
32 maintworld 2/2017
ASSET MANAGEMENT
7
Mining machinery is diverse
sometimes simple like a convey-
or, other times more complex as
in the case of a reclaimer. The condition
monitoring requirements in the mining
world are therefore quite diverse and
frequently not simple.
Airborne ultrasound, sometimes us-
ing a parabolic dish pointing out of the
window of a pickup is a very quick and
reliable means of inspecting the condi-
tion of a conveyor especially if it is
12km long.
There is more than the usual amount
of slow-speed equipment in mining,
which is often in critical operational
roles. Ultrasound is perfectly capable of
listening to bearings rotating at even less
5
There are many electrical than 1rpm and still providing valuable
Figure 1 systems involved in the min- diagnostic information.
ing industry from DC to HV. In Finally, there is the need to consider
many cases dust is a major contributor the condition of machinery that is itself
to failure. One of the key problems asso- moving like shovels, or that are moving
ciated with the build-up of dirt, dust and violently like vibrating screens.
moisture on the surface of components A critical bearing rotating at 24rpm
is Tracking. shown in Figure 2. was found to have
The ceramic insulator pictured in failed during an ultrasound inspection.
Figure 1. failed because it was covered The bearing had recently been replaced,
in dust which was causing the tracking. so it was relatively new. Unfortunately,
On-condition cleaning using ultrasound the replacement bearing was not quite
to identify the presence of the tracking is the correct one and was undersized for
used to eliminate failures by optimising the load requirement. Very quickly, the
cleaning procedures. new bearing disintegrated.
Figure 2 The mining community in South Af-
8
rica is also leading the way in the adop- Over-lubrication is quite an
tion of ultrasound as a safety-screening established tradition in the min-
tool to protect electricians working in ing world grease that bearing
substations. A small ultrasound kit is until I can see the grease coming out of
located at the entrance to the substa- the sides. In a previous article in this
tion and there is a series of assessment magazine, I reported how one mine
measurements to be performed in order successfully reduced its lubrication con-
to provide approved safe access to the sumption by over 95% whilst improving
building and proximity to the panels reliability against all established meas-
inside. This approach is undoubtedly ures in place.w
saving lives by providing a higher level of So there you have it. One technology,
safety in the work environment than can Ultrasound, used in either airborne or
be provided by flameproof or arc-flash contact mode to identify problems in
clothing alone. 8 major problem areas in mining. Con-
servatively, in the last decade the sav-
6
Tightness testing of the air ings that customers have achieved must
intake systems of the large be well beyond $25M. Time for you to
diesel engines in haul trucks start?
2/2017 maintworld 33
ADVERTORIAL
THERMOGRAPHY
IN PHOTO QUALITY
THERMAL IMAGING CAMERAS localise Geometrical resolution plays a crucial Its the inner values
defective heat exchangers, detect faults role in the maintenance and inspection that matter...
in the electronics, overloaded mechani- of plants in the electronics industry. When compared with conventional de-
cal components or increased energy Here, detector resolutions that are too tectors, the most powerful models of the
consumption. The visual representation low can quickly cause areas of concern VarioCAM HD have a higher geometri-
of the temperature distribution provides or potential danger points to be over- cal resolution, resulting in a significant
a quick overview of the plant condition looked. Sources of error can be ruled increase in the field of view and about a
directly on-site. This can be documented out with thermal imaging cameras such 40 percent higher range. This is notice-
over long periods of time with the aid as the VarioCAM HD and VarioCAM able in practice: As a whole, fewer pan-
of thermal images so that maintenance HDx. With the extensive measuring, tilt-movements and single frames are
measures can be set at the optimum objects such as photovoltaic systems can necessary, which shortens and stream-
time. This increases efficiency, produc- be thermographically measured as well lines operating time on site.
tivity and profitability and ultimately as objects requiring safety distances, like Even small temperature differences
extends the service life of plants, too. high-voltage installations. can be detected with certainty thanks
34 maintworld 2/2017
ADVERTORIAL
2/2017 maintworld 35
CMMS
Connectivity and
Interoperability:
No value without reliability
A central challenge posed by Industrie 4.0 and the Industrial Internet of Things (IIoT)
is the secure, standardized exchange of data and information between devices, ma-
chines and services across different industries.
AS EARLY AS April 2015, the Reference service-orientated architecture (SoA) manner, which enables them to partici-
Architecture Model for Industrie 4.0 helps humans understand the things pate monetarily in big data applications
(RAMI 4.0) recommended only IEC more quickly and easily, because they and data analytics.
standard 62541 OPC Unified Architec- offer services and describe their un- To exchange the data, OPC UA com-
ture (OPC UA) for implementing the derlying meaning. The subject of SoA bines two mechanisms to implement
communication layer. In November is nothing new in the world of IT. Now, various scenarios:
2016, the Industrie 4.0 Platform pub- however, it extends all the way to the A client-server model, in which OPC
lished a checklist for classifying and things themselves. This is where OPC UA clients use the dedicated services of
advertising products as Industrie 4.0 UA comes into play, providing the frame- the OPC UA server. This peer-to-peer
Basic, Ready or Full. To comply work for industrial interoperability. approach provides a secure and con-
with the Industrie 4.0 communica- Machine and device manufacturers de- firmed exchange of information, but
tion criterion, even the lowest category scribe the object-orientated information with limitations regarding the number
requires the product to be addressable of their systems and define the access of connections.
over the network via TCP/UDP or IP and rights along with integrated security A publisher-subscriber model
to integrate at least the OPC UA infor- features. where an OPC UA server makes config-
mation model. urable subsets of information available
When it comes to information mod- Security Built-In by Design to any number of subscribers. This kind
elling, many small and medium-sized Germanys BSI (Bundesamt fr Sicher- of broadcasting mechanism provides an
companies tune out, because they com- heit in der Informationstechnik, or unconfirmed fire and forget-style ex-
pare OPC UA with other protocols like Federal Office for Information Security) change of information.
MQTT and assume that it has limita- published the results of its in-depth OPC UA offers both mechanisms,
tions. We often hear questions like, OPC security analysis of the OPC UA specifi- but the more important benefit is that
UA cant communicate directly with the cations and a selected reference imple- they are decoupled from the actual
cloud, can it? mentation in highly positive terms: An protocol. TCP and HTTPS are available
First of all, every equipment and ma- extensive analysis of the security func- for the client-server model, while UDP,
chine manufacturer already provides an tions in the specification of OPC UA con- AMQP and MQTT are available for the
implicit information model with data in- firmed that OPC UA was designed with publisher-subscriber model. As a result,
terfaces (via various protocols). Humans a focus on security and does not contain the question of OPC UA or AMQP or
have learned to adapt to the computers systematic security vulnerabilities. MQTT doesnt matter from the OPC
way of thinking documenting what As a result, the machine builders Foundations perspective. Since the
the bits, bytes and hex codes mean. This keep full control of the data, i.e. they can smallest microcontrollers may not have
new world full of devices capable of a distribute it in a targeted and controlled enough resources to implement fully-
36 maintworld 2/2017
CMMS
Any product
being advertised
as Industrie
4.0-enabled must
be OPC UA-capable
(either integrated or
via a gateway). The
checklist also stresses
the information
modelling property German Electrica
l and April 2017
of OPC UA. Electronic Manufac
turers Association
fledged OPC UA, the device can offer its device vendors sales, but not supporting For example, manufacturers can bill for
data over MQTT or AMQP in an OPC the OPC UA standard will definitely de- their barcode or RFID reader on a per
UA-compliant manner, making it easier crease them considerably. scan basis while the data being read or
to integrate it at the other end. After all, scanned never leaves the factory.
agreeing on an information model and Trend: SoA
what the data means is the key to achiev- Most of the industry-specific informa- Trend: Chip-based OPC UA
ing the concepts of Industrie 4.0. tion models developed so far are no OPC UA will continue to be integrated
longer based on the exchange of bit/byte into ever-smaller devices and sensors.
Trend: Information models properties, but rather on SoA services Todays smallest OPC UA software so-
OPC UA provides secure transport of with complex parameters. An OPC UA lutions for industry with limited (but
data via diverse and expandable proto- client that does not support any methods read-able) functionality require just 35
cols. But who defines the datas mean- for this purpose or complex parameters KB of RAM and 240 KB of flash memory.
ing? Other associations like AIM for the will be increasingly hampered in its Now that the first chips with integrated
auto ID industry (RFID readers, scan- communication with OPC UA servers. OPC UA have hit the market, OPC UA
ners, etc.), VDMA technical groups for An RFID reader offers no bits to activate can make further in-roads into the world
injection moulding machines, robotics a read/write command, but instead uses of sensors. As a result, OPC UA applica-
or machine vision and 35 other VDMA methods that can be read by humans: tions are already extending from the
industries already define their informa- ReadTag, WriteTag, and KillTag, among core area of automation into other areas
tion in OPC UA servers in the form of others. OPC UA is ideal for SoA imple- like industrial kitchen appliances.
so-called OPC UA companion specifica- mentation, which is why the German
tions. For an equipment supplier, meet- Commission for Electrical, Electronic & Conclusion
ing this type of industry standard does Information Technologies (DKE) lists OPC UA has already become the de-facto
not automatically mean they become OPC UA as the only SoA solution. standard for the automation market
exchangeable, as each manufacturer can and Industrie 4.0. OPC UA is covering a
offer their own special services on top of Trend: Service-to-Service growing range of communication sce-
the standard. Intelligent devices should OPC UA provides consistent scalability narios, which makes it increasingly dif-
definitely be able to support multiple from the sensor to the enterprise IT ficult for suppliers to justify proprietary
information models simultaneously level, making a significant impact on solutions. Products will increasingly
for example, the dedicated functionali- the automation pyramid. While this differentiate themselves based on the
ties of an injection moulding machine, pyramid will continue to exist for the features of the device itself or of external
in addition to the models for energy factorys organizational structure, OPC services, not the interface. In the future
data or MES interfaces. To reduce the UA bypasses the communication pyra- we will see rapid growth in the informa-
engineering effort, the importance and mid entirely. The devices can deliver tion models of additional industries, as
availability of such industry-specific and data, either directly or in parallel, to OPC UA is the preferred platform of the
multi-industry information models will the PLC, MES, the ERP system or to the worlds largest ecosystem for interoper-
increase rapidly in the future. OPC UA cloud level. This is where suppliers see ability.
may not directly increase an industrial opportunities for new business models. www.opcfoundation.org
2/2017 maintworld 37
SECURITY
SECURITY
SOLUTION
Helps Battle
Illegal Mining
Mining companies across the African continent have been coping
with the problem of illegal mining for years. In their battle
against artisanal miners illegally entering their properties and
compromising site safety, they have been unsuccessfully looking
out for security solutions that are both reliable and affordable.
DAVID MONTAGUE,
Sales Director EMEA/
Security, FLIR Systems, David.
montague@flir.com
38 maintworld 2/2017
SECURITY
2/2017 maintworld 39
CYBERSECURITY
Digital Automation
System Maintenance
and Cybersecurity
the Perfect Partnership?
In the last years, the number of cyber-attacks has
ROBERT VALKAMA,
increased dramatically. In light of this, it is not Senior information
security consultant at
surprising that organizations are increasingly looking Nixu Corporation,
to invest in cybersecurity. robert.valkama@nixu.fi
40 maintworld 2/2017
CYBERSECURITY
substantial benefits for maintenance. bling superfluous software) is primarily an exception is detected in the same
For example, asset and configuration intended to reduce the relevant systems way cybersecurity-related alarms and
management, systems hardening and attack surface, but can also have the ad- events are generated. This is especially
security monitoring. ditional benefit of removing potentially beneficial in multivendor environ-
Asset and configuration manage- faulty software from the relevant sys- ments where automation systems from
ment is one of the corner stones in tem. This, in turn, reduces the need for several suppliers are used. In multiven-
cybersecurity. Without accurate knowl- unnecessary maintenance. dor environments the different systems
edge of the environment, it is very Cybersecurity monitoring is another may be monitored separately, using
hard to reliably secure it. A complete, function that can be easily utilized in their own diagnostic tools, but a com-
accurate and up to date documenta-
tion of the systems in an easily acces-
sible (queried) format is required in
order to be able to rapidly respond ASSET AND CONFIGURATION MANAGEMENT IS ONE OF
and investigate the potential impact of
newly discovered vulnerabilities on the
THE CORNER STONES IN CYBERSECURITY.
protected environment. A traditional
blue print type of documentation is maintenance. These tools focus on mon overview is not available. Depend-
usually not sufficient for this, as they keeping track of an automation sys- ing on the environment and personnel
dont contain important and needed tems network traffic and scouring its size the monitoring may also be limited
information of the digital devices like, logs. They are intended to identify ex- to post incident resolution, or extin-
used software and versions, firmware ceptions or changes, which means that guishing fires as some may refer to it.
versions, configuration information they can also be configured to moni- With good and high quality monitoring
etc. This information is essential for ex- tor maintenance-relevant informa- the maintenance of the digital assets is
ample when conducting vulnerability tion, combining and centralizing two shifted towards a preventive mainte-
assessments, but it will also streamline separate functions. Monitoring tools nance mode, where incidents are iden-
fault diagnosis. can be configured to generate mainte- tified before they cause any process
System hardening (removing or disa- nance-related alarms or events when disruptions.
Discover Discover
the hiddenthe hidden
treasure intreasure in
Maintenance
Maintenance
There is value hidden in every maintenance organization. All companies have the potential to further improve, either by reducing
costs, improve safety, work on the lifetime extension of machinery or by smart maintenance solutions that improves uptime. The
question is where maintenance managers should be looking to nd these areas of improvement and where they need to start.
You will nd the answer to this question at Mainnovation. With Value Driven Maintenance and the matching tools like the VDM
Control Panel, the Process Map and our benchmark data base myVDM.com, we will help you to discover the hidden treasure in
your company.
ARE PRACTICALLY THE SAME: ENSURING ERROR-FREE A better way to protect your sys-
tem against cybersecurity attacks is
PRODUCTION AND SAFETY. to connect it to the company network,
and route all the needed connections
through a dedicated access point, allow-
How to Improve Cybersecurity related to the operation. I would not ing the management and monitoring of
When considering automation systems recommend putting too much effort into remote connections and ensuring that
from a cybersecurity standpoint, one assessing the probability of different existing cybersecurity controls are not
challenge stands out above all: their long events, but concentrating on the poten- bypassed. Continuous monitoring will
lifecycle. IT systems have a lifecycle tial consequences and their acceptance. also help you identify remote sessions
of around five years, whereas automa- I.e. if a potential cause of a cybersecurity from your automation systems vendor
tion systems have an average lifecycle attack utilizing a remote connection and changes made through these con-
of around 25 years. What this means in could lead to an extensive equipment nections to the systems configuration.
practice is that even though todays au- failure or jeopardize employees health In other words, monitoring tools can al-
tomation systems suppliers work on im- or life, a strong argument can be made to so be utilized for contract management,
proving the cybersecurity of their own make some changes to remove or mini- allowing you to monitor the suppliers
systems, it will take up to 25 years for mize the risk. The risk assessment needs actions, and for configuration manage-
these built-in security features to per- to be a multi domain task, performed in ment, allowing you to verify whether
meate the entire manufacturing indus- co-operation by cybersecurity experts, planned changes have actually been im-
try, and at that time, some of the security process engineers, safety engineers and plemented.
controls may already be obsolete. maintenance engineers. All in all, those working with automa-
However, it is possible to substan- Regardless of the outcome of the risk tion systems should deepen their mutual
tially improve the cybersecurity of assessment, here are some recommen- collaboration. This is especially true for
automation systems, even though some dations what should be done. maintenance and cybersecurity profes-
vulnerabilities might still remain. It is Consider securing your plants safety sionals. Solid cooperation ensures that
also important to acknowledge that all automation solutions or safeguards, of all aspects required for safe and stable
assets are not equally important, and course provided that they are digital. production are taken into account. From
that the security posture of a system can With these I mean functions intended early planning stages to decommission-
be substantially improved by making for protecting employees, production ing and dismantling, modern cyberse-
good engineering decisions for example equipment and the environment against curity must be considered throughout
on the architecture and functionality al- accidents or hazards. Where possible, an automation systems lifecycle. When
location. you should isolate protective automatic considering digital cybersecurity solu-
It is also recommended to perform a systems or functions from the operative tions, I would recommend checking that
risk assessment. The purpose of the risk automation systems. This should also be your organization has access to the latest
assessment is to identify the potential a basic rule when designing new control and most comprehensive know-how in
plant vulnerabilities and critical systems systems. the cybersecurity industry.
42 maintworld 2/2017
CONDITION MONITORING
44 maintworld 2/2017
2/2017
CONDITION MONITORING
ment, valves, hydraulic applications and sibility issues: will a ladder or man-lift be to process equipment, and finally to the
steam traps. needed for any steam traps that are over- condensate recovery systems.
However, when it comes to steam trap head and out of reach? Are any steam When the inspector is at the steam
inspection, one technology alone cant traps located in areas where Lock-Out- trap, before testing with ultrasound, it is
do everything. World Class maintenance Tag-Out procedures will need to be fol- recommended to take temperature read-
and reliability programmes utilize mul- lowed? Are any of the steam traps inside ings with a simple spot radiometer first.
tiple inspection technologies when it of hazardous areas where Intrinsically Not only will the temperature let the
comes to inspecting the assets that they Safe instruments will be required? inspector know if steam is coming to the
are responsible for. Visual inspection, trap or not, but the temperature can also
temperature measurement, and ultra- Testing Steam Traps with be used to estimate the steam pressure.
sound should all be used during a steam Ultrasound If the temperature of the steam trap is
trap inspection route. To make it easier to manage the data for cold, the inspector should check to make
reporting and inspection purposes, it is sure that the valves are open or if the
Planning for Success recommended to break the inspection trap has been taken out of service. If it is
Before beginning any steam trap inspec- areas into zones. Follow a logical pro- warm/hot, then the inspector can note
tion, it will be helpful to think a few gression from steam production, steam the inlet and outlet temperatures and
things through that will help to make the use, and condensate return. Start at the proceed to test with ultrasound.
survey successful. First, walk the area to boiler, then to the steam distribution and The most important item that the in-
identify and tag every steam trap. The distribution branches. Then, proceed spector will need to know is which type
tag should include a number, and all in-
formation should be noted, such as the
manufacturer, type of steam trap, orifice
size and the purpose of the steam trap.
AN ULTRASONIC STEAM TRAP TESTING PROGRAMME IS
This information can then be entered AN EASY, QUICK, AND ACCURATE WAY TO IDENTIFY
into data management software, such as
UE Systems Ultratrend DMS.
PROBLEMS IN STEAM TRAPS AND THE OVERALL HEALTH
While walking the area, it would also
be a good idea to note any possible acces-
OF THE STEAM SYSTEM.
On-board sound
recording and sound
analysis can be very
useful for steam trap
inspection
Information models of different branches are mapped onto OPC UA to make them interoperable with
integrated security.The OPC Foundation closely cooperates with organizations and associations from
various branches:
TM
Verband fr Automatische
Datenerfassung, Identifikation und Mobilitt
OPC D2A01Y7
OPC Unied Architecture
Interoperability for Industrie 4.0 and the Internet of Things
EUROPE
mation
IT meets Auto
IoT
Industrie
4.0
osted by
May 30 /31 h
M2M
th
Micros o
Download of
Technology brochure:
opcfoundation.org/
resources/brochures/ www.opcfoundation.org
HSE
Overview of Dutch
Working Conditions in 2016
In 2016, occupational health and safety policies in Dutch companies were
more widespread than in 2014, but still below the level of 2007, according
to an Overview of Dutch working conditions 2016 -report.
THE OVERVIEW of Dutch working condi- (RI&E) and 75 percent of companies of medical and other care for people
tions 2016 report is published biannu- with a contract with an occupational with a work-related condition (1.4 billion
ally presenting the current state of the health and safety service provider and euros). Together, these costs added up to
working conditions and work-related the provision of information and train- 8 billion euros, thereby accounting for
health of employees and of health and ing. Given that large companies employ more than 20 percent of all the costs of
safety policies at companies in the Neth- the majority of employees, these policies sickness absence, disability, and medical
erlands. involved 83 percent and 96 percent re- care for those in work in 2016.
According to the report, in 2016, there spectively of the Dutch employees. The
was an increase in compliance with the increase in health and safety policies in Employee health is stable
key provisions of the Working Condi- 2016 may be related to economic devel- The report shows that the level of health
tions Act by companies, such as having opments. experienced by employees has been
a contract with an occupational health stable for many years: 82 percent of
and safety service or service provider, Accidents and Diseases employees described their own health
sickness absence-related policies, a pre- Work-related accidents and diseases as being good or very good. Meanwhile,
vention employee, work meetings with lead to high costs. These costs include less than 2 percent assessed it as poor to
employees and provision of information costs of paying salaries to employees very poor. Of all employees, 18 percent
and training. Compliance percentages who are absent due to work-related said they have a long-term or chronic
varied between 45 percent of companies issues (4.7 billion euros), disability condition that restricts the ability to
with a risk inventory and evaluation benefits (1.9 billion euros), and the costs perform their work optimally. This
48 maintworld 2/2017
HSE
50 maintworld 2/2017
Successful reliability
programs have one thing
in common
a strong condition monitoring team.
As a condition monitoring
professional or manager, you may be
faced with starting a condition monitoring
program or tasked to get a program back on
track. There is a lot to know; multiple technologies,
monitoring practices, analysis, fault-reporting
and trending, not to mention proper corrective and
routine maintenance to reduce the likelihood of faults from
reoccurring. Where can you start? Where can you learn what you
need, that is economical in respect to time and expense? And where
DBOZPVHFUUSBJOJOHUIBUJTQSBDUJDBM
HPJOHGVSUIFSUIBOKVTUMMJOHZPV
XJUIGBDUTBOEHVSFT
iLearnReliability
[Condition Monitoring]
TO PROVIDE PREMIER
SOLUTIONS FOR
PLANT MAINTENANCE.
Laser shaft and geometric alignment
Portable vibration analysis and eld balancing
Online Condition Monitoring
Worldwide training, services and support