Maintenance Task Selection
Summarised by : Sandy Dunn
Webmaster, Plant Maintenance Resource Center
This 4 part summary from the plantmaint Maintenance discussion forum discusses alternative approaches to Maintenance task selection, including RCM, PM Optimisation (PMO), RCMCost, and others - it also touches on Total Productive Maintenance (TPM)
Go to Part 2 of this discussion
From Peter Ball
Ron (who lurks no more),
Welcome to the fray.
Just a couple of basic questions:
1. Are you aware of any other RCM "Packages" that conform to the requirements of SAE JA1011;
2. If so, can you name them?
3. If there are others, then why have you come out so strongly in advocating RCM11, which after all said and done is only the title of a book.
Please do not mis-understand me, but industry IS confused with this never ending 'tussle over RCM and it's many variations, most if not all fail in some respect or other. Hence the need for TQM in many instances.
From Ron Doucet
To Peter, who is definitely not a lurker,
Sorry about the delay in responding to this but it is a busy time of the year
To answer you first question regarding other SAE JA1011 compliant RCS, yes there are others namely the US Naval Air Command's NAVAIR 00-25-403 "Guidelines for the Naval Aviation reliability Centered maintenance Process" and the British Royal Navy with its RCM-oriented Naval Engineering standard NES45, these process have remained true to the process that was named RCM by Nowlan and Heap in 1978. I know a lot of so called RCM processes out there and I have compared them and none except those that I mentioned
For those of you who don't know, the evolution of RCM started in the 1960's to address the reliability of aircraft in an attempt to keep them airborne. The aviation industry started to focus on maintaining what the aircraft did, not what they were. In other words they were maintaining the functions of the aircraft. With so much success and enabling planes to be reliable and at minimizing their maintenance costs, the US Department of Defence commissioned a report to aid them in maintaining their aircrafts.
In 1978, a report was prepared for the US DoD describing the current state of the process that the airlines were using. the report was written by Stanley Nowlan and Howard Heap of United Airlines. It was entitled "Reliability-Centered Maintenance", or RCM.
From general consensus and from what I have seen, no other comparable process exists for identifying the true, safe minimum of what must be done to preserve the functions of assets. This includes not only maintenance tasks but also operator tasks, training, process changes and physical redesigns.
To answer you second request, yes
For your third, when it is all said and done, RCM II is not only a title of a book (worth reading by the way) it is also the name of the RCM process. Unlike all of the other RCM's, whose focus was to take the sound work of Nowlan and Heap, (if they read it) and try to make is faster, RCM II was the attempt at industralizing the process such that it could be applicable to any industry not only aviation. One of the major changes was not in how the process is applied but to actually make it more rigorous such that environmental consequences could be managed.
In 1978 when the Nowlan and heap report was published, the environment was not the biggest issue and for aircrafts, the focus was on "maintaining the desired functions of the aircraft, to take off, fly and land, not on environmental consequences after a crash.
I have come out strongly in favour of RCM II because other so called RCM processes either skip steps , do not embody the functional approach, or are merely preventive maintenance optimizing tools. As for the latter, they are focused on improving an already existing maintenance program and not on determining what the maintenance program should be. True RCM, determines the required activities to ensure that any physical asset continues to deliver its desired functions in its present operating context.
The problem with all of the "expedient or quickened versions" that I have seen is that they are not safe. Keep in mind, some so called RCM process do not even resemble one bit the original RCM. Of those that do, steps that were taken out in the streamlining effort without knowledge of the consequence.
I was even at a conference where this vendor of an RCM type process said "With (this type of RCM) you can identify 85% of critical failures in much less time than classical RCM" (read RCM II).
The question I had to ask was, what are we going to do with the other 15% of critical failures?
Apart from the fact that RCM II is not slower than any so called other RCM process, can we stand up in court after someone has died and justify the ignoring of critical failures for expediency of a process.
Here is something to think about as you are choosing how you will determine you future maintenance programs. As there are a lot of Australians on this site, this should be close to you hearts:
After the Longford disaster in September in the State of Victoria, the State parliament on November 13th 1998 added a new section to the State of Victoria evidence Act of 1958 which reads as follows:
"19D. Legal professional privilege
(1) Despite anything to the contrary in this division, if a person is required by a commission to answer a question or produce a document or thing, the person is not excused from complying with the requirement on the ground that the answer to the question would disclose, or the document contains, or the thing discloses, matter in respect of which the person could claim legal professional privilege.
(2) The commissioner may require the person to comply with the requirement at
hearing of the commission from which the public, or specified persons, are
excluded in accordance with section 19B
In other words, when all is said and done, this amendment suspended
attorney/client confidentiality for the purpose of Longford and future official
I will leave the decision up to you, as for me, I have chose a sound, audible
and defensible process. I would hate to be in front of an inquiry telling people
that I tried to save a few buck on a process which documented that it followed
the 80/20 rule.
Anyway that's all for now.
I hope I have addressed you questions and comments
From Steve Turner
I would like to make some responses to your email (attached) which made
some comment on PM Optimisation. PM Optimisation is a relatively new tool
but has been found to be better than RCM for many operations. Through this
email, I would like to explain why and clear up some of the
miosunderstandings that I think you may have picked up around the tracks.
I hope that the group may be interested in me providing some information on
This email is going to be fairly long but I hope interesting... so if your
going to have a read, give yourself five minutes.
If you want to go to a website about PMO, visit www.pmoptimisation.com.au
There are two detailed explanation papers which you can down load. One is
about applying PMO in a mature organisation and the other is about using
PMO in the design phase of a project.
My source of information
The information that I am using is based on 12 years of consulting in the
area of RCM and other maintenance improvement tools.
For four years in the early 90s I was a Licensed RCM II trainer and
facilitator. For the past four years, I been working along side Strategic
Industry Research Foundation (now known as SIRF Roundtables Ltd
(www.sirfrt.com.au)) and 8 major multinationals from a variety of capital
intensive industries. My work has resulted in a well proven tool to
improve asset management. The work commenced with evaluating the approach
taken by the North American Nuclear Power Industry.
What is PM Optimisation?
PM Optimisation is a method of maintenance analysis specifically designed
for existing plants. It has also been used to assist in the design phase
of new plants where there is experience in dealing with ones that are similar.
The methodology that we have developed is called PMO2000. PMO 2000 has
nine steps as follows:
Step 1 Task Compilation
Step 2 Failure Mode Analysis
Step 3 Rationalisation and FMA Review
Step 4 Functional Analysis (Optional)
Step 5 Consequence Evaluation
Step 6 Maintenance Policy Determination
Step 7 Grouping and Review
Step 8 Approval and Implementation
Step 9 Living Program
Download a paper from the website for more details of these steps.
In general terms PM Optimisation starts from the premise that mature
organisations must be doing reasonably well to be still in business so the
current PM program can't be that bad.
Having said this, we commonly find that most of the industry problems
emerge because of insufficient resources to complete the program as well as
the corrective maintenance requirements. We therefore have a strong focus
on human as well as asset productivity and have designed a method which
retains the necessary levels of coverage and detail, but is done in a
vastly improved time frame.
The next premise is that in mature plant, most of the likely failures have
been experienced or have been managed effectively by the current preventive
maintenance program. Those potential failures that have not occurred may
not be economically preventable anyway (depending on cost of down time) as
the cost of failure averaged over the life would be low compared to the
cost of prevention.
Essentially what is done by using PM Optimisation is to gather in one list
all of the formal and informal maintenance tasks done by all those involved
in operating and maintaining the equipment. (It is worthwhile noting that
we find on average 50% of the PM completed is done on an informal basis
through the initiative of tradesmen or operators.)
The next step is to define the failure modes that each task is meant to
prevent or detect.
We then sort and filter all the failure modes and add to the list, those
which have happened in the past and are not receiving any PM and also add
to the list any failure modes that are considered likely to happen. We
often do a bit of a Hazop at this point using PID or other technical
references including any manuals.
Once failure modes are listed, the functions lost by each failure can be
listed. This step is optional and rarely done in practice. To most
uninitiated people this is heresy however, we believe that a functional
analysis is, in most cases, not good value for money. The example I would
like to use is that of a family car or a mobile phone. I would like you to
ask yourself what value you would get by spending a day fathoming all the
functions of either before working on the maintenance analysis. Having
said that, there have been times when the group needs understand the
functionality to come to a satisfactory outcome and this is why we say this
step is optional.
Once failure modes have been established then, consequence analysis and
task determination proceeds in line with recognised RCM decision logic and
the rest is no different to RCM.
In summary, if I were to explain in one paragraph the differences between
traditional RCM and PMO it would be this:
PMO and RCM methods are both methods of maintenance analysis aimed at
defining the complete maintenance requirements of assets however, the
central difference is the way that the body of likely failure modes is
generated. RCM starts from a zero base and arrives at a list of failure
modes by defining functions and functional failures whereas PMO generates a
list of likely failure modes through an analysis of the current PM program
and adding to that, the list of failures that have already happened and
those which, on examination of the design are considered likely to happen.
This process is on average six times faster than Classical RCM (I can
provide the reference paper if anyone is interested). In addition this we
believe that PMO2000 obtains a maintenance coverage very close (98%) of
that which you would obtain using classical RCM methods.
Comments on Ron's propositions.
"I have come out strongly in favour of RCM II because other so called RCM
processes either skip steps , do not embody the functional approach, or are
merely preventive maintenance optimizing tools. As for the latter, they are
focused on improving an already existing maintenance program and not on
determining what the maintenance program should be"
We believe that the whole point of PMO is to determine what the maintenance
program should be. Perhaps you mean something else that I have missed.
"The problem with all of the "expedient or quickened versions" that I have
is that they are not safe."
You may like to expand on this Ron as there are no reasons provided.
However I would like to make the following points:
First, the PM Optimisation program used in the Kewaunee Nuclear Power
Station (USA) for many years has been assessed by the equivalent or our
Regulator General as a major strength. This came after the company
abandoned the use of RCM.
The second point is this. Lets say you have 30 hazards in your plant, one
of them is receiving no PM, three are receiving the wrong PM and the others
are being maintained properly. Can you wait five years to find out the
problems or do you want to find out straight away. If you use PMO2000 you
will have these under control in one year, if you use traditional RCM it
will take you six.
Further to this, in my ten or so years of facilitating RCM analysis, I have
put about 1 in every 200 failures in the hazard category. Of these, only
once have I ever felt the RCM team had uncovered a potential hazard that
was not receiving any PM.
My rough calculations tell me that the benefit of RCM over PMO2000 is the
one new hazard found in 15,000 failure modes. 15,000 failure modes would
consume 15,000 manhours with four people at a workshop running at a rate of
15 minutes per failure mode. Based on a 6 to 1 rate of analysis (PMO/RCM),
identification of the hazard has cost 12,500 extra manhours. I would
suggest that you find other ways to use that labour to make a contribution
to site safety.
The third point is that the greatest safety issue surrounding maintenance
is not the lack of a good PM Program but a lack of resources to do it and
the big worry is the lack of resources to do the corrective action. You
mentioned the Esso disaster. I would like to add to your comment that in
this disaster a significant contributing factor was a lack of corrective
maintenance not a poor preventive maintenance strategy. I read in the
inquiry findings that it was known that there were many items in the plant
that were not serviceable and the unserviceability of some of these items
contributed to the scale of the disaster. This to me is an indication that
Esso were carrying far too many failures.
The problem we often find with industrial companies is that when
maintenance people finally get the plant, the corrective work consumes the
time allocated and many times the PM is missed. We call this the vicious
cycle of reactive maintenance. In short, high backlog of corrective work
and high levels of unplanned failures leads to missing PM - Failed
equipment almost always has a higher priority than PM. Missing PM
increases the rate of unexpected failure and so the cycle gets worse. We
have evidence of this through data collected from some of our clients.
A safety conscious maintenance engineer needs to get out of this vicious
cycle, get on top of the backlog and get back into PM. To do this he needs
to create productivity in his people (eliminate over servicing, task
duplication and ensure that all tasks are done at the correct frequency)
and find a tool that will get him that productivity and control quickly.
If this is the road to a safer environment then PMO2000 will give you a
much better safety profile than RCM will.
"I was even at a conference where this vendor of an RCM type process said
this type of RCM) you can identify 85% of critical failures in much less
than classical RCM" (read RCM II)..............
The question I had to ask was, what are we going to do with the other 15% of
There is quite a likelihood that the presenter was me and if this was the
case, I have obviously failed to get the point across correctly. What I
intend to say at conferences is this:
When doing a zero based RCM according to the principles of Nolan and Heap,
one is attempting to find all the failure modes that are reasonably likely
to occur. This results in somewhere around 20 PM tasks for every one
hundred failure modes (if you do RCM without skipping any steps). Thus 80%
of the analysis time has resulted in "No Scheduled Maintenance". This
provides little value if the objective of the analysis is to generate an
effective PM Program. With PMO2000 we get less than 10% of analysis
resulting in "No Scheduled Maintenance" hence the analysis time is more
productive and the coverage six times greater for the same analysis time.
The big point is that we get the same maintenance program out the other
end, not 85% or 80% of it.
Finally, I personally regard RCM according to the paper written by Nolan
and Heap as the greatest contribution to reliability engineering throughout
history. I have a copy on my bookshelf.
The point that escapes most people however, is that RCM was originally
developed as a design tool rather than as a tool for mature organisations
with a mature maintenance schedules.
Therein lies one of the greatest contradictions to RCM being used in mature
plants - the intended function of RCM was for use as a design tool yet
people have for a long time, been trying to use it as a tool for improving
maintenance in mature organisations.
For all those RCM Gurus out their, at Step 1 of RCM, functional analysis,
we find need to redesign the process of RCM because it is being used in a
manner that was not intended and is not capable of achieving a satisfactory
level of performance because it takes too long.
If you have gotten this far, grab a Fosters from the fridge and have a beer
on me for Christmas :-)
Don't forget to download the papers from the web at www.pmoptimisation.com.au
From Tony Ramshaw
A very interesting note and very much in line with my experience.
It also fits in with a previous note of mine on the forum that you need to
split out the real integrity items and treat them differently.
I used to work for Shell from the Hague who developed a simplified RCM
process S-RCM many years ago. It works on many of the principles you
describe and has been well proven over the years. It also consumes around
20% of the effort of RCMII and as a result gets used and implemented, so the
benefits are quickly delivered. Again as you describe it involves the
technicians so you get real practical input plus they learn the economic
value of maintenance. It also has a strong focus on the business benefit of
doing preventative maintenance, if there isn't any don't bother.
Head of Maintenance - Kvaerner Australia
From Stephen Young
Interesting comment Tony.
I guess it comes down to identifing all of the failures that could ruin your
business or kill peolpe - or do you just a skim across the top to find
some of the failures.
My information is that some within Shell are now thinking the S-RCM approach
was not such a wonderful idea after all and in fact, has left them more
exposed to risk and litigation.
I understand the issue of trying to find a quick solution in a world
demanding immediate results, but that is a project management and analysis
priority issue, not an analysis process issue.
From Tony Ramshaw
Thanks for the comment, agree with he general sentiment.
If people in Shell are really saying that S-RCM is not such a good idea that
is news to me. I don't believe you are exposed to risk if you look at
integrity items separately and that is what Shell do using RBI and IPF.
The whole process is of course the intelligent application of business risk
management. If you don't have the tools or experience to do this then you
need to apply something like RCMII and FMECA rigorously. Even that does not
fully solve the issue because unless you pick your teams carefully with the
right experience you can still miss things. A very experienced facilitator
is required and many of the RCM facilitators only know how to apply the
method but do not have the experience to know if the answer makes sense.
From Peter Ball
Hi there Steve,(and Sandy 2)
You seem to be on the right track with your PM Optimisation.
Interesting that you also have described "Conventional" RCM, but it has
integrated with Classical" RCM in terms of Long time and High cost. Several
RCM Vendors would dispute this, especially those claiming that Streamlined
is better than Classical.
I am very curious to know the root origin of the seven basic questions which
appear to define the only RCM approach, and have copied them direct from SAE
4.1 RCM-Reliability-Centered Maintenance
5. Reliability-Centered Maintenance (RCM)-Any RCM process shall ensure that
all of the following seven questions are answered satisfactorily and are
answered in the sequence shown as follows:
a. What are the functions and associated desired standards of performance of
the asset in its present operating context (functions)?
b. In what ways can it fail to fulfil its functions (functional failures)?
c. What causes each functional failure (failure modes)?
d. What happens when each failure occurs (failure effects)?
e. In what way does each failure matter (failure consequences)?
f. What should be done to predict or prevent each failure (proactive tasks
and task intervals)?
g. What should be done if a suitable proactive task cannot be found (default
To answer each of the previous questions "satisfactorily," the following
information shall be gathered, and the following decisions shall be made.
All information and decisions shall be documented in a way which makes the
information and the decisions fully available to and acceptable to the owner
or user of the asset.
I have spent many hours searching RCM documentation but without real
success. These include amongst numerous others: MSG-3, Mil-Std-2173
(Mil-Hdbk-2173) both cancelled in favour of Navair 00-25 403 (current).
These questions would appear to be the origin of RCM (Classical?) but who is
to rule that there are other approaches, and decision diagrams, which can
also be considered valid. Yes I have a copy of RCM11 and Yes ... the
questions are there , virtually identical to those in SAE JA1011.
I am not suggesting that an RCM approach can be based only on Pareto
Analysis, as alluded by Ron.
On your marks, ready..Set...GO!
From Kim Matulich
With all this talk of litigation it's amazing we don't have the company
legal eagles doing reviews of their equipment strategies.
Stephen, in an economic downturn how do you convince an organisation to
implement a thorough RCM process in comparison to say a TDBU approach to
strategy development? I'm genuinely interested as a lot of maintenance
decision makers I have met look mainly at the tangible returns (minimum
cost, minimum project duration) rather than the projected expected returns
of carrying out RCM.
From Ron Doucet
Steve, thanks for the rather long reply to my comments on RCM and RCM processes.
I will address all of your concerns to what I wrote, and, back up some of my
points for clarification. Given everything you wrote it will take a little time.
From reading some of Steves information on his site, his PMO process, which he
does not call RCM, has more RCM in it that most streamlined processes that call
themselves RCM. Steve though, called it what it is, a PM optimization process,
From Jose Duran
RCM II is only another approach to RCM. for me RCM is RCM , the main
differences between versions is how you apply it.
I particularly think that RCM across all the assets is too expensive. We
believe in the RCM risk driven (criticality, cost/risk optimization, etc.),
where it is only worthwhile for high risk systems. There are several
companies that have faced those costs yet and now they are performing a
streamlined version of RCM
From Ron Doucet
Jose, actually RCM is a defined process with discrete application steps. It is
not an approach. If you change how you apply it, ie take out some steps, apply
it differrently, then by default it is not RCM any more.
Not to get hung up on this but RCM is a process that was defined by Nowlan and
Heap in 1978. If you change that process, good or bad, (not to get in a debate
right now), then it is no longer the same process and therefore no longer RCM.
It is amazing how alot of people critisize "clasical" RCM to varying degrees yet
they all want those 3 letters in their process names. If someone has a better
or different process then call it a different name and let it ride on its own
If you take the vodka out of a screwdriver (mixed drink involving vodka and
orange juice), its not a screwdriver anymore..........its orange juice.
More details on that later
From Stephen Young
If the changes to Victorian (Aust) legislation (aftermath of Longford) go
through parliament then we will have the legal departments taking a very
active interest in how we maintain our assets.
On your second point, there is the myth that RCM is expensive when compared
to the modified RCM approaches. It is just that - a myth and we often see
rigorous RCM analyses achieving ROI's measured in weeks and in some cases,
days. The key is commitment, proper training, good project management and
sensible selection and prioritising of the assets analysed.
We consider it is better to spend your effort doing selected analysis that
you can defend, and gaining all the benefits of that rigorous analysis, than
putting your effort into an analysis which you know is not catching all
failure modes that could bite you and which is not maximising the benefits.
(A bit like repairing an item of kit - you can do a 'dash across the top
band-aid' repair which gets you out of immediate trouble or, a more thorough
job which restores the item to it's original capability. Often the time
difference between the two approaches is minimal - but the benefits huge.)
It is at the end of the day, a judgement call, but we believe it is better
to do the job well rather than spend a similar amount of time to achieve
less benefit. There will be times when a very quick analysis may be
expedient but to us, it is better to do a higher level analysis to identify
where you are at risk rather than change the analysis process.
Selling the idea to your board should be on the same basis as selling a
capital expansion or upgrade of plant to the board. You describe the need,
the business case, the risks, the benefits to the company (and particularly
the board), the project plan, the outputs and how you will measure your
success. It works for our clients.
From Trevor Hislop
When setting up a quality PM system in an established plant, it is usual
practise to use a small area of the plant as the trial-horse. In this way,
it is very easy for all to see the results and it makes it so much easier to
then "sell the idea" to senior management, bean counters et all.
Can you do this with RCM,????
From Ron Doucet
Yes you can. That is the approach we chose here at the Iron Ore Company of
AS you all know by know we went with RCM II for reasons that i have already
mentionned and will mention further in by future responce to Steve.
To make it short.
Chose a pilot (actually chose too many as beginners in 1996 but that's
That pilot was an automated rock crusher.
It was producing at 50 to 60% capacity. We need approx. 750,000 tonnes of
crushed rock per year to maintain our haulage roads. It was delivering
between 400 and 500 thousand at $6.00 per tonne!
Did an RCM analysis as a pilot. Pilots take 2 to 4 times longer than normal
due to the newness of the facilitators and training (that's OK as long as you
know it up front)
Total cost of the pilot was approximately $80,000 to perform the analysis.
Analysis stayed on the shelf for a while.
We are still producing at $6.00 per tonne
Someone suggested that we close the crusher and buy the stuff at $3.00 per
I suggested that they listen to their people who gave a lot of effort in
using RCM as a process to determine the maintenance and operating
requirements for this crusher.
We implemented one suggested redesign, the operating procedures and the
For 18 months prior to implementation we were producing at $6.00 per tonne.
One month after implementation, yes one month, we were producing at $2.00 a
tonne which at the time I was tracking it, was sustained for 16 months( and
like the energiser bunny, its still going and going and going...)
End result: $3.0 Million dollar savings on unit cost per year, year after
An unexpected $750,000/year savings on direct maintenance costs.
So for a total of a one time $80,000 investment (today it would cost only
$20K to $30K because we are much faster) we have been saving $3.75 million
dollars per year since 1998.
So we want to reduce the cost of doing the RCM to get a bigger return? Is
this not enough?
Another quickie, again using real RCM.
Now we are more experienced so things are going the way they should.
RCM analysis on a dewatering filter of which we have 26.
Total analysis, using that long RCM II process took 13 half days with 6
Total investment in doing the RCM analysis $13,100.00 (yes only 13 thousand
Total benefits based on changing one mtce task from fixed interval to
Total benefits (unit costs) for 26 filters was $8 Million/year
No benefits on direct maintenance costs as these are all small motors etc.
Benefits were realised by decreased unplanned operational consequences.
13 half days = $13,000 cost = $8.0 million per year benefit
What are we trying to speed up again? How fast do we want to go?
And to answer your question,
the company listened
One more note.
As you can see from above, the lion's share of benefits were unit cost
decreases through increased the capacity of the assets. This was achieved by
first determining what we wanted the equipment to do and then by detailing
the most appropriate failure management policy that would decrease or
eliminate the operational consequence. Processes that focus mainly on
reducing direct maintenance costs will miss the big picture and the bulk of
the benefits by orders of magnitude of 1000s. RCM II of course focuses on
delivering the maintenance tasks, operating procedures, identifying the
required training and detailing the required redesigns that will enable your
asset to perform at its desired level and do what it is supposed to do.
I meant this response to be shorter but I guess its tool late now
From Kim Matulich
Which Victorian (Aust) laws are you referring to? What recommendations from
the Longford enquiry are the catalyst for these changes?
I have a copy of the enquiry and would be keen to highlight your references
for future justification.
From Jim Turner
Ron, dollars are good but you gave no indication of how it was done !!
Lets get the before and after, where did you find savings?
Bar talk is always good but lets get to the bottom line, e-speak is time
is money etc.
From Ron Doucet
For the automated rock crusher:
The after and before:
750,000 tonnes per year produced at $2.00 per tonne instead of $6.00 per tonne =
yearly savings of $3.0 million
Mtce savings came from one of the suggested redesign basically to go from
hydraulic drive to electric and improved reliability. (decreased rate of
failures due to maintenance and operational procedures.) those were $750,000 per
year off the bottom line.
For an item to be maintainable it first has to be able to deliver its required
function. In our case the hydraulic drive couldn't. This is one of the reason
that a maintenance policy should be based on functions. this ensures that you
maintain what the equipment does and not what the equipment is.
For the filter, the savings resulted mainly from going to on condition filter
By doing that:
the clothes were only changed when they had to
it also allowed for planning and scheduling of the changes at times when the
operational consequences were decreased
And it decreased the consequences of filter clothes being operated when they
were inefficient which not only decreased throughput but also cost us extra in
additives to maintain quality.
The $8.0 million dollars savings is all in the extra capacity enable through a
different failure management policy for the cloths.
Even though we determined on condition maintenance tasks for the motors etc.
which we previously ran to failure, You still have to replace the motor so there
were no big savings there except for a bit of overtime.
We make our money by passing 12 million tonnes of slurry through the filter so
that we can make our end product, iron ore pellets. The cost of the operational
consequence is 10,000 times the cost of the failed components. Your bang for the
buck is to focus on doing the right things, using the right process and enabling
your asset to deliver its intended function. The biggest cost benefits are in
decreasing unit costs. That is how most of us make a profit out there.
I hope this answers your question.
In terms of how the analysis was done, it followed the RCM II process from start
to finish and then through implementation of the findings, as I said before, one
redesign, the mtce tasks and the operating procedures.
Now I hope I have answered your question.
From Sandy Courtalin
I think you should have closed the crusher and buy stuff @ $3.00 per tone.
A quick calculation shows that it is more economical to subcontract than
to carry on running your crusher. In addition you save all your maintenance
costs linked to the crusher and you can obliged your supplier to reduce its
selling cost each year within a partnership contract.
About the suggested redesign, what was the cost to up-grade the crusher?
Personal thought about RCM....:
Sometimes we think that improvement are due to the using of a method like
RCM. But at the end the problems pointed out and the solutions found were
generally well known by maintenance operators or more generally by the whole
operators. Maybe they just didn't have the possibility to report about it
a suggestion scheme for example. So in my opinion, before rushing in
and magic methods", the maintenance manager should ask himself:
Am I well using the resources currently available?
From: Trevor Hislop
Great reply Ron
Thanks for down-to-earth practical information. It's real world stuff !!!!
Not dreaming in an ivory tower !!!
I am sure most forum members would like to hear of more practical examples
with "real" figures from RCMMMMMMM !
From Mick Drew
I think what Ron has done is outlined his recommendations from the RCM in
the language of the business- this is essential no matter what method is
used. If maintenance recommendations are aligned to achieving the business
goal - in this case dollars per tonne thruput - then the recommendations
have greatest chance of being supported by management.
In many cases however recommendations for changes to maintenance plans,
redesigns, ideas from suggestion schemes etc are not connected in a logical
way to the impact of failure and so the business impact. My experience in
industry with maintenance improvement methods - RCM, TDBU, FMEA, has been
that there is a scepticism from senior management about the investment of
time and resources in methods which still come down to an individuals
"judgement". Unlike RONS experience, some management would prefer to leave
the RCM documentation on the shelf and simply take the lowest cost option.
It normally then takes a charismatic champion to do battle to support the
Now, I have been using the RCM method coupled with computer simulation. This
means the improvement tasks chosen can be simulated over a lifecycle, and
the benefit to the business estimated in Cost of failure, failure frequency,
unavailability, safety, environment and operational impact. This allows
alternative scenarios to be compared- do nothing, versus "Optimum" versus
"Optimum minus 20% cut in budget". The difference in expected results can be
evaluated in hours. Management can treat the RCM program as any other
business improvement process. Typically, the alternative sceanrios are
presented to compare the cost of implementation, the cost of failures, and
the risk of safety, environment and operational incidents.
This process overcomes the problems encoutered when "expert opinions" are
challenged, when management are after business impacts and the get
"promises" of best practice or believe in RCM its good for you after all it
worked on planes.
The output of computer simulation is also useful for resource planning as it
forecast manning and spares requirements for each scenario. Because today's
knowledge is captured in quantifiable parameters it is possible to update
from future equipment history so the analysis outlives the individual.
From Ron Doucet
Like I said, it was the manager who thought he could buy it for $3.00 a
Where we are that is highly unlikely.
Secondly if we were doing everything right we would be producing for close
$1.00 a tonne so I think were are going to keep operating the crusher. Even
$2.00 a tonne I'll take the $750,000 savings over that of a contractor.
By the way, the modification costs of the redesign were including in the
unit cost and the drop from $6.00 / tonne to $2.00/tonne still happened in
month and was sustained. It even amazed me.
With respect to your comments about "revolutionary and magic methods" All
did was apply a process, determine the maintenance requirement and implement
magic). RCM is not revolutionary (been around for almost 40 years) nor
It is organised logic with sound foundations. The people who applied the
were the operators and the tradesmen themselves.
With respect to the people on the floor already knowing what to do and what
wrong, I agree that you can brainstorm and get a lot of improvement ideas.
the other hand if people have old fashioned views and ideas about the
maintenance function, i.e. maintenance fixes things, then you could actually
yourself in trouble. You also can not show a process nor audit trail when it
comes to showing due diligence. As you know showing due diligence and
becoming ever more important as society's tolerance for industrial disasters
With no process somebody says, put a bigger "something" as a solution and
end up causing bigger problems elsewhere. These problems could have safety
Anoth risk is when people come up with a solution of installing protection
(outcome of HAZAOPs) yet the maintennce of that protection is ignored.
have fun justifying that tactic if an accident occurs due to protection not
working and the maintenance was not addressed.
Many times the maintenance people only show up when something is broken so
do not know what an impending failure sounds or feel like.
Part of the process in and RCM II analysis is the change the way people
about maintenance and then it allows that changes thinking process to be
develop the right approach for maintaining that asset.
People who have seen the inner workings of an RCM II session will agree with
that one of the most spoken phrases during an analysis, which is done in
not alone, is "hey, I didn't know that" from the operator, "hey, I didn't
that" from the maintainer and "hey, I didn't know that" from the supervisor
And Holy @#%$!, I didn't know that could happen!
I could go on about the benefits of groups using a well founded process and
about the risks of doing otherwise. I will talk about that later, now I have
go home for supper.
Oh ya, one more thing. Most maintenance people out there believe that if we
the effort up front in planing and scheduling that the overall effectiveness
will be increased. Our choices are unplanned maintennce vs planned
We chose planned.
The same applies to the gathering information. We can do it in a planned or
unplanned fashion. Unplanned is an adhoc brainstorming or problem solving
session which generates a lot of ideas.
The Planned approach is first educating people on the process and then using
that process to get the right ideas out. The up front investment in using a
process will make the overall exercise actually shorter, much like what
does to a job, and it will ensure that the right ideas come out, much like
the right job right.
Here is a story
In actual fact, I had told some people a couple years ago that they should
the RCM II process to address a potential safety hazard which had resulted
actual work refusals.
Two years later after many, many,many periodic meetings, problem solving
sessions, consultant reports, involving lots of people, the problem was not
Than I got a phone call, they told me to go ahead and try to solve it using
7 meetings later, 6 people per meeting, a good facilitator and the problem
corrective actions were determined. The people on the floor could not solve
before. Now the people on the floor solved it using the process.
I hope I addressed some of your points.
So again a lot of people on the floor have good ideas, but it does not hurt
actually get those ideas out, and more, by using a systematic approach. The
process will actually allow you to use your resources well so that you can
answer yes to "Am I well using the resources currently available?
Now I'm going home.
Talk to you all tomorrow
From Sandy Dunn
I am not Stephen, but as I understand it, the new Victorian legislation
which has been introduced has developed a new regulatory framework for
workplaces which potentially represent "major hazards". At present, this
consists of around only 50-60 workplaces in the State of Victoria, most of
which are in the petrochemical or chemical industries.
The legislation requires these companies to develop a set of policies and
guidelines (which have to have certain features) for identifying potential
major incidents and minimising the risk of occurrence. These guidelines
must be reviewed, and approved, by the appropriate regulatory authority. I
would imagine that there will be significant penalties for those
organisations that either do not comply by submitting appropriate
guidelines, or who fail to follow their approved policies and guidelines.
This largely came about because of a realisation that the existing
Occupational Health and Safety legislation, while appropriate for managing
"minor" events, such as injuries and even fatalities, was not appropriate
for managing major catastrophic events. The Longford enquiry established
that the Longford explosion occurred despite the fact that Esso had in place
a health and safety system that was considered to be one of the best in the
world. Anecdotal evidence suggested, however, that the focus on site was on
relatively minor safety issues (for example, employees were regularly
encouraged to replace safety glasses that had scratched lenses), while at
the same time, gas leaks (however minor) were generally treated by employees
as relatively routine events.
Incidentally, Esso had intended (and planned) to conduct a Hazop on the
plant that exploded for several (I think 3) years prior to the explosion,
but this analysis had been regularly deferred, and had still not been
conducted at the time of the explosion. The Longford Royal Commission
concluded that, had this Hazop analysis been conducted, it is highly likely
that this would have identified the potential risk, and that appropriate
actions would have been taken to minimise this risk. Given this judgement,
it is highly likely (in my judgement) that the regulatory authority would
approve any structured Hazard identification and review process (including
Hazop, RCM, RBI and others) proposed by these few organisations that are
subject to the new legislation.
As an aside, let me add to Steve Turner's comments by saying that I, too,
was an RCM II consultant and facilitator for almost seven years, and during
that time assisted many of Australia's largest organisations to implement
the RCM II process. During that time, my experience was that, for every ten
organisations that started to implement RCM II, only one ever implemented
the process on anything other than a "pilot project" scale. This failure is
not because of any failing on the part of the organisations that tried RCM,
nor, I hasten to add, was it due to any failure on the part of the
consultant involved! I have recently come to the conclusion that, in
contrast to the position that is put forward by John Moubray, Ron Doucet and
others from the RCM II religion, the major problem is that RCM II (and by
definition RCM) has NOT been adapted sufficiently to meet the needs of
industry outside the airline industry.
Its major failure is in the way that it treats risk at a macro level. The
reality is that, in the Airline industry, the consequences of major failures
of significant components on aircraft are the same - 10's to 100's of people
die, and the photographs are on the front pages of most major newspapers.
In different industries, however, the consequences of major failures of
significant equipment items vary significantly - in the petrochemical, oil
and gas industries, there is a chance that perhaps 10's of people die. In
the chemical industry, perhaps 100's or 1000's of people die. But in a
semiconductor manufacturing plant, the major consequences are economic, and,
in comparison, relatively minor. Furthermore, even within those industries,
the failure of different equipment items have totally different levels of
risk associated with them - in the petrochemical plant, for example, failure
of a furnace could be catastrophic, but failure of the rainwater sump pump
far less so. Because of these differing consequences, at a macro level, the
amount of time and effort that should be devoted to identifying risks and
possible failures should be appropriately different. However, the RCM
process treats all such systems the same way - the same amount of rigour is
required. I have heard it said (even by John Moubray himself) that one
cannot justify applying RCM to all equipment items - some equipment items
have such low impact on business risk that the effort required to perform
RCM analysis on them is greater than the potential benefits. My question,
then, is this - if we are not to use RCM to develop PM programs for these
equipment items, what process should we use?
In my opinion, the PM Optimisation process outlined by Steve Turner, IS RCM
adapted to suit the real needs of all industries. It can be as rigorous as
is required to identify all potential failures, hazards and risks (a la
RCM), but it equally can be quickly applied to give 90% of the benefits of
RCM at a fraction of the cost, where the time and effort to perform a
rigorous RCM analysis cannot be justified. It uses the RCM concepts (such
as PF Interval, the Decision Diagram, etc.), but does not strictly follow
the RCM process as outlined in JA 1011. If you want to get semantic as a
result, and say that it is not, strictly speaking RCM - then that is fine.
In that case, I believe that we have a lot to thank the team that developed
JA 1011 for. First, it clearly identifies those processes that claim to be
RCM-based, but clearly are not. Secondly, it now allows us to see what RCM
really is - a rigid, overly rigorous (in most cases) process that may be
applicable to airlines and some other high risk equipment items in selected
high risk industries, but is complete overkill in most situations, in most
For the record, I now assist organisations to implement the PM Optimisation
process, with great results.
From Stephen Young
Thank you for replying on my behalf, but the legislation you mentioned is
not what I had in mind when I penned my comment in my email of last week.
The state governments of Victoria and Queensland are both considering
legislation to deal with "Industrial Manslaughter (Vic)" and "Corporate
Culpability (Qld)" as both governments are concerned their current
legislation does not adequately deal with industrial incidents causing death
or serious injury. Victoria is leading the way after the Longford incident.
Importantly the concept of aggregation of negligence is introduced which
allows the aggregation of actions and omissions of a group of employees and
managers to establish that an organisation is negligent.
Both governments have made it clear that if managers and/or a management
system fails to prevent workplace death or serious injury, then the
responsible manager and/or management team is likely to face criminal
Should the legislation proceed, penalties of over $500,000 and 7 years
imprisonment are proposed.
From Jose Duran
Now I see more people agree with me. I am working with RCM since about 6
years. I think it is not worthwhile to be applied in all the assets. it
would take a lot of time and efforts, and the return on low criticality
systems is so low. Despite that if you decide to apply it over all the
assets, you should at least make a very good criticality study to do first
the first. I mean than ,many RCM efforts have failed because the pilot
studies were done in low criticality systems, of course low ROI after it.
In process industry the assets could be a lot. A big mining company could
have invested US$10.000 million in equipment (or more!). Ok, who dare
performs FMEA in all of them!
We have developed an approach called RCM (+), we use RCM with several
modifications to fit it into the non aeronautical industry. But we use that
approach only for high and medium risk systems. We also developed another
tool called RCM reverse to deal with low criticality systems (about 40%)
that give you results quite good in a fraction of time and efforts that
typical RCM studies.
From Jean-Francois Ducrot
Jose, could you explain what you mean by "reverse RCM"?
From Jose Duran
RCM reverse is a way we deveoped to study low criticality systems. It allow
us to use RCM logic to verify if the current maintenance strategies worth.
It is easy and fast, but it should be performed by really expert RCM
From Sandy Courtalin
I think there is a misunderstanding about my thoughts about maintenance
methods. I just tried to say that in my opinion there is no more than 15%
of the world wide industry which has ever used method like RCM once.
I know that RCM appears about 40 years ago but it was principally used
by people who didn't care about the maintenance costs (army, aerospace)
because it was about safety issue in a first instance.
I would like to say that still in my opinion, achieving "maintenance best
practice" doesn't mean applying RCM, TPM or CMMS. Most of the
time, companies don't have the skills or the resources to use these tools
or don't have the budget. I think specially about small companies where
it is common pratice to have a manager appointed with two hats "maintenance
and production" and quite often the maintenance is considered as the fifth
wheel of the trailer.
So, when they decide to consider the maintenance, you can't arrive
and say "you shoud start an RCM approach". Maybe they need tips
about service organisation, planning, shut-down... to achieve "best
maintenance practice" to comply with their commitments. The maintenance
policy must be defined from the strategy of the company and then
find solutions in relation with the goals to reach.
Hope you have a better understanding of my thoughts.
From Trevor Hislop
I agree with you. I have worked with many companies to bring them out of a
"fire-fighting" (reactive) mode and start them on the way to more effective
maintenance. In most cases, they do not need a CMMS ,or RCM, or whatever,
and they just cannot afford it anyway. I usually teach them the basics of a
sound, simple, Preventive Maintenance program that they can start in one
small plant section, e.g. air compressors. I also get them to give ownership
of what is happening directly to one or two of their maintenance staff. The
results are amazing! When proven, it is very easy at very little cost to
expand activities to other plant areas.
Remember Nero fiddling while Rome burned ??? Well that is what happens quite
often when big buck maintenance enhancements are suggested.
From Peter Reed
I believe it is beer shouting time again.
Trevor/Sandy, what you are saying is the way it SHOULD be. Too many times
we see companies spending huge amounts on TLA's only to wonder why all their
maintenance woes have not magically disappeared. While the AP's etc of the
world have great products they are still just that, products, not mystical
methods for taking your problems away. It takes getting your hands dirty
and talking face to face to solve the real problems with our respective
We see time and time again where a proactive format, for doing work, yields
results far exceeding any computerised program but we always seem to revert
to these more expensive programs and I have theory on this:
The people, who sell these programs are very well educated and exceedingly
knowledgable. They are also very good speakers and salesmen. Joe Bloggs
the Tradesman is a mug (in the nicest possible way) and only has his hands
on experience and gut feel to express his knowledge. Put both of these in
front of a Production Manager, the Salesman with all the right words, the
overheads and the glib ability to make you believe, the tradesman with his
dirty overalls, course language and ability to draw mud maps, and who do you
think will get the funds on the day?
What we need is a group of managers who can combine the results of the fancy
suits, the flash presentations, the dirty scrawls on the back of a coaster
and the vivid language and come up with a strategy that suits the needs of
the plant AND the business.
I have been reading, avidly, the RCM debate, on this site, and actually took
some of the E-mails to our trades staff to have a read. Most of these people
had a glance through and then said "What they really mean to say is....."
From Ray Beebe
One valid reason for a company or plant taking up activities indicated in
this long discussion as TLAs is surely to shake up complacency, such that
real improvement results when people are jolted out of their "comfort
zone"? Hence such actions as re-organisation, shuffling managers around,
etc. How long should someone stay in the one position to avoid
complacency? My own longest stay in the one post was 4 years, the shortest
On a related note, mentions of "productivity" grate on me. I heard a
senior manager describe how his power plant was run by such a very small
crew of people that "they would all fit into this room". Nobody was
brave enough to ask him about the numbers of contractors who performed the
maintenance work, or those who, on contract, operated and maintained the
water treatment plant, the ash handling plant, etc!.
The only important issue is surely how much is the total cost, however
arranged, or perhaps better still for international benchmarking, the total
number of manhours (Personhours?) used in performing whatever work you are
Season's greetings - and I look ahead to enjoying more of these debates!
Ray Beebe at Monash University (now, academic productivity - that is
From Peter Ball
I love it!
I Just Love 'IT',
Good story came out of WW11, origin was a certain European General. He
considered that the Allies had a real problem which affected their
behaviour. If they were confronted with any situation not clearly defined in
basic simple terms, they had to have a Meeting and consult the appropriate
Manual. Europeans (on the other hand) were capable of thinking through their
problems immediately, because of better initial basic training. Thus ...
battlefield meetings were not often necessary.
We may see a parallel here with RCM which is, after all, nothing more than
plain old "common sense". An attribute that was somewhat lacking in the
country of it's origin, at the time.
Just go searching for MIL-STDs' pertaining to RCM, Logistics support, FMEA,
etc. They raise them, work them, and cancel them with great heaps of
paperwork. No small wonder the average workforce comments are ... !#@%^&***
Go figure ... I love it. (Beer and WINE also)
From Jean-Francois Ducrot
I'm sorry, Jose, but I have still some problems with your reverse RCM logic.
It's obvious RCM is a very heavy approach for low criticality systems, and
another way of doing could be very interesting.
But I can't understand how you make RCM simpler for those systems. Could you
give further explanations on your methodology?
Go to Part 2 of this discussion
Copyright 1996-2009, The Plant Maintenance Resource Center . All Rights Reserved.
Revised: Thursday, 08-Oct-2015 11:53:48 AEDT