Steve,

quote:

this happens then that happens which triggers this reaction and that human decision under that specific circumstance". These are the chains that cause disasters... RCM does not even come close to resolving them.

You are quite right. RCM cannot resolve how people decide after a sequence of events is initiated. You go on to say, again quite rightly that
RCM will reduce your exposure to catastrophic events but it will do this by reducing the vicious cycle of reactive maintenance

You have hit the nail on its head. My point as well is the 'how', not the 'what' of RCM. The way RCM works, as far as the prevention of event escalation, is by making sure that the barriers to escalation actually work on demand. Many of these barriers have hidden functions. RCM attempts to expose these villans by detecting them in time and ensuring that they do work, or at least that enough of them do work on demand to break the sequence of dominos.

I explored this subject and found two cons for rcm as quoted below. Do you agree or disagree with these views?

"Failure data drive reliability-centered maintenance (RCM) programs required for 15-30% of the equipment which can benefit from RCM and we support total productive maintenance (TPM) efforts for 60-80% of the equipment in most operating plants."

http://www.barringer1.com/reliability.htm

"The purpose of RCM is "to determine the maintenance requirements of any physical assets (EQUIPMENT) in its operating context."1 This is accomplished by answering seven questions about the equipment in order to determine what type of maintenance strategy to employ for the asset. RCM provides a flow diagram that tells you what type of maintenance to use based on the answers to the questions. By answering the seven questions all of the potential modes of failure are uncovered and a predictive maintenance strategy is devised to mitigate the consequences of the failure based on the criticality of the failure mode. In RCM, these failure modes are identified as the root cause(s) of the failure. This is where the main difference lies.

The purpose of RCA is "to uncover the underlying reasons (root causes) why an event (not just equipment but any type of event) is occurring so that the necessary steps can be taken to eliminate the event in its entirety." This is accomplished by analyzing the modes (the point at which RCM stops). RCA uses a logic tree that stresses verification at every level. The advantage is that the actual root causes that are uncovered are facts that have been derived from the verification process. The comparison between the two programs is striking -RCM is driven by preventive maintenance strategies while RCA is driven by maintenance prevention strategies.

It should be clear that the difference between RCM and RCA is that RCM treats the symptom while RCA finds and corrects the cause."

http://www.maintenanceresources.com/ReferenceLibrary/ezine/rcmvsrca.html

Hm.. I think everyone is writing his own maintenance bible on the internet these day...

Comparing RCA with RCM !!

If the RCA unveils that the cause of the premature failure is that some $%^& idiot hammered a bearing with a 10 lbs sledge hammer, you can not blame the type of maintenance.

The true root cause could reveal:
1) this individual was never trained
2) his boss was never trained
3) the individual is a psychopat
4) his boss did not recognize it
5) management brings in any lunatic
6) maybe the head honcho is lunatic and everything must be done yesterday because the company is loosing money.

This is not a maintenance problem, this is a company management problem.
And the "poor" maintenance organization getting blamed.

RCA is a tool that can be used also for maintenance

quote:

It should be clear that the difference between RCM and RCA is that RCM treats the symptom while RCA finds and corrects the cause."

In my previous example, RCA if done properly can find the cause. But the correction ??
Many people will consider the do nothing action

Evaluating if the Do-Nothing is a valid alternative or not, should be first in the solution finding process.

Quote:
The Relationship between RCM and TPM

The original precepts for RCM (refer page 3) were developed for the aircraft industry where 'basic equipment conditions' (no looseness, contamination or lubrication problems) are mandatory, and where operators (pilots) skill level, behaviour and training is of a high standard. Unfortunately in most manufacturing and mining operations these 'basic equipment conditions' and operator skill and behaviour levels do not exist thus undermining the basis of any RCM application.
For this reason, the application of TPM as a company wide improvement strategy is highly advisable to ensure:

'basic equipment conditions' are established; and
'equipment-competent' operators are developed
Before attempting a full blown RCM analysis or a partial RCM approach following the basic RCM process. Failure to do this in an environment where basic equipment conditions and operator error are causing significant variation in the life of your equipment parts will block your ability to cost effectively optimise your maintenance tactics and spares holding strategies.

The other key difference between RCM and TPM is that RCM is promoted as a maintenance improvement strategy whereas TPM recognises that the maintenance function alone cannot improve reliability. Factors such as operator 'lack of care' and poor operational practices, poor 'basic equipment conditions', and adverse equipment loading due to changes in processing requirements (introduction of different products, raw materials, process variables etc) all impact on equipment reliability. Unless all employees become actively involved in recognising the need to eliminate or reduce all "losses" and to focus on 'defect avoidance' or 'early defect identification and elimination' failures will never be cost effectively eliminated in a manufacturing or mining environment.

Conclusion

It should be acknowledged that a TPM implementation is not a short-term fix. It is a continuous journey based on changing the work-area then the equipment so as to achieve a clean, neat, safe workplace through a "PULL" as opposed to a "PUSH" culture change process. Significant improvement should be evident within six months, however full implementation can take many years to allow for the full benefits of the new culture created by TPM to be sustaining. This time frame obviously depends upon where a company is in relation to its quality and maintenance activities and the resources being allocated to introduce this new mind-set of equipment management.Unquote

http://www.plant-maintenance.com/articles/RCMvTPM.shtml

In my company we developed a maintenance strategy on increasing HSE, financial, reliabilty and availabilty performance, where we use RCA and RCM.

RCA is being used as a feedback loop for increasing availabity. Monthly a TOP 10 is made of the unplanned and unwanted disturbances. Together with production the priorities are set and the maintenance engineers use RCA together with the required specialists to find the Root Cause of the problem and not the solution. After the root cause is found, actions are defined to prevent the disturbance from happening. The actions are agreed with the production and maintenance manager and executed.

RCM is used as a feedforward loop. To get out of the firefighting cycle, preventive maintenance is defined based on preventing unwanted disturbances to happen (based on FMEA). I totally agree that this is a tough and long process. Our goal is first to focus on the 20% of the equipment that causes 80% of the trouble (Pareto). In this way you will reduce the length of the process by focussing on what is really important.

In my opinion, if you really want to structurally increase you HSE, financial, reliabilty and availabilty performance there are no real shortcuts. Well managed RCA and RCM processes and the involvement of all key-players in plants is essential.

regards, Erik

quote:

The other key difference between RCM and TPM is that RCM is promoted as a maintenance improvement strategy whereas TPM recognises that the maintenance function alone cannot improve reliability.

Josh - I dont quite buy that one
If you want to have your operators understand the importance of thier role in maintenance - have them participate in RCM / PMO.

Hello Josh,

RCM is a structured approach in determining the maintenance requirements of an asset in order to fullfull its function.

I do not say disadvantage but rather difficulty in implementing is that setting up the team is easy and this is a selection process which will compose of the most experience maintenance and operators of the asset. But to do the analysis requires a series of meetings from 2 to 4 hours depending upon the complexity of the system being analyzed and this will be the difficult part, just dont expect 100% attendance everytime most specially the operator and after the analysis had been completed still we can say that it is still subjective, I recall that we perform an analysis on one of our substations and after the implementaion it ran very well with very few failures but one rainy season, a lighting struck and this was way way out of the failure mode lists and was unexpected.

Performing RCM is a tedious process specially when writing down the effect in paragraph form from 20 to 60 words per failure mode, and everthing starts from scratch, here is where PMO advantage is since according to Steve Turner that it will start with your current maintenance activities being performed. However, when you have completed and implemented the analysis, then you can sleep better at night.

Another difficulty we encountered is that it will take a lot of guts to implement such an analysis. Have you heard about the Add on PM syndrome where the activities on PM continues to multiply again and again, specially when the people that add up to your PM lists of activities will be your customers, CEO, Audits, Quality Control and other people around passing by your equipment and notice something ???

Rolly Angeles

Some commercial content below:

Quote

quote:

Mike, I read the John Moubray book which represents one view of RCM. A collective view (based on concensus hopefully) is better I guess and thus this post.

quote:

Regardless of whether you suppor thte RCM 2 approach or not, Moubrays book remains the best text I have read on RCM in the marketplace today. (And I have read most of them)

There is no doubt that anyone who wants to implement RCM should read Moubray's book. You'll need its information to defend or propose whatever you want to do as people seems to fall into several very opinionated camps on the RCM subject.

Another book that many people find as a good reference is RCM:Gateway to World Class Maintenance, by Mac Smith and Glen HInchcliffe.http://www.reliability-magazine.com/bookstore.htm

Finally a newer book on RCM is available, Reliability Centered Maintenance – Implementation Made Simple by Neil Bloom. Bloom is a classical RCMer. However, he does believe that the process can be greatly shortened and the amount of resources needed reduced by huge amounts. There's an interview on the reliability magazine web site in the top right corner with Neil Bloom if you'd like to learn more.

The magazine is also sponsoring a reading club, and this book will be the first one we tackle. So, if you're looking for more RCM info, want to argue about it and have an excuse to read this RCM book, then http://www.reliability-magazine.com/reading_club.htm

quote:

Originally posted by Vee:
If 'streamlined' or turbo or jet RCMs can do the same job cheaper, let us go for it, BUT not if they can miss any sleeping tigers. Oh yes, we do need to know ALL the credible failure modes, not pick and choose the juicy ones only. We only need to miss one really important hidden failure to suffer badly.

-

Vee:

I honestly have a problem with accepting this comment. I do not believe that Classical, RCM 2, or any other "non-streamlined" RCM will absolutely catch all "sleeping tigers". So by this statement I am led to believe that if RCM in any form can miss one of these "sleeping tigers" than we should not perform it at all?

This whole "streamlined" arguement is getting old. I am not just referrring to what you have stated. There are a lot of flimsy and unsubstaniated claims out there with no real definitive evidence. I have read Moubray's article and many more and none of them identify any specific "streamlined" RCM processes and give undisputed proof of how they do not meet the so called standard. They are all general statements any many of them are differences in point of view.

The true fact is that even if only one improvement a day is made that makes the company better than the day before. It is the mentality of continuous improvement that needs to be pushed and not the mentality that if your not doing it my way than you shouldn't do anything. I am not implying that you are saying this but the tone and direction points this way.

There are a lot of people who have been categorized as providers of "Streamlined" RCM who have provided wonderful services and improvements to companies and to discount that is shameful. The real fact is that where cost is a factor in everything some of the processes are just giving the consumers (companies) what they are asking for. A faster and cheaper way of making improvements. In other words they are listening to the voice of the customer. I cannot tell you if they are right or wrong because in truth only the customer can make that decision. Obviously, there are a lot of customers that are happy with the accused "streamilned" RCM providers because a lot of them are still out there working.

No RCM analysis is perfect and that is part of the reason why we have sustainment so we can improve off of what we have done.

On another note, I do want to say thank you though for all of the great inputs and advice you have provided to the forums. They have been helpful. This is just one point of contention I have with a lot of the posts and accusations that are made on the forums about RCM.

Thanks

Robbie,
[/QUOTE]

quote:

I do not believe that Classical, RCM 2, or any other "non-streamlined" RCM will absolutely catch all "sleeping tigers". So by this statement I am led to believe that if RCM in any form can miss one of these "sleeping tigers" than we should not perform it at all?

I am afraid I have not made my point clear, so I think it has led to a misunderstanding. I have nothing against 'streamlined RCM', and am not an 'RCM 2' fan. What I think is important is that the process must be structured, thorough and examine all credible failure modes.
RCM processes that use pick-lists are perhaps OK for the expert but not so with novices, as some failure modes may be missed. Similarly, those that go straight from Functions to Failure Modes are flawed. There are some software packages that call themselves RCM-this or RCM-that, but have no resemblence to anything that Nowlan & Heap proposed.
To revert to your comment, I agree fully with you that no RCM process is guaranteed to catch all failure modes. But some will catch only 80%, some others 95% and yet others 99.5%. Faced with a Texas City type scenario, if you are defending the case in court, which one would you rather have followed?
I dont want to get into emotional hype,as the earlier question may imply, but I am advising caution to many forum members who may pick 'the cheapest' thinking they are getting value. It is horses for courses; the higher the hazards in your industry, the more picky you should be. In other posts I have suggested that 50-60% of reliability gains from just doing some simple things. If you dont need the RCM effort, dont do it, but if you need it, make sure you have one that matches your industry needs.

I agree with that. I think that everyone needs to do thier homework before picking an Rcm provider out there. I also think that companies need to be extra careful and not send someone to school and have them come back to the company and try to go it alone. Mentoring is an important part of RCM. I am sorry if I misunderstood your direction. You are a veteran in this arena and I know you can understand the frustrations with some of the zealots out there who preach thier way is the only way. Thanks for the clarification.

Robbie,

quote:

I also think that companies need to be extra careful and not send someone to school and have them come back to the company and try to go it alone. Mentoring is an important part of RCM.

Personally, I do not support RCM activities that are not team-based, with a rep from Ops, Mech and Instr Maint, and if required, Insp, Rotating Specialist, Process Technologist on a part time basis. So a solo job by one trainee ain't good enough in my book!
An outside RCM expert as a Facilitator is a valuable resopurce; I know it costs money, but he/she can keep the RCM team along the right path. That is good value.
These are personal opinions which you may accept or reject.

quote:

Originally posted by Vee:

Personally, I do not support RCM activities that are not team-based, with a rep from Ops, Mech and Instr Maint, and if required, Insp, Rotating Specialist, Process Technologist on a part time basis. So a solo job by one trainee ain't good enough in my book!
An outside RCM expert as a Facilitator is a valuable resopurce; I know it costs money, but he/she can keep the RCM team along the right path. That is good value.
These are personal opinions which you may accept or reject.

I agree. That is what I am saying. Too many companies send someone to school and try to have them run an RCM program internally without any real guidance. This is a bad practice. As they gain experience than they can manage an internal RCM program but without experience they are setting themselves up for failure. I think the best practice is as you say to have an RCM professional as a facilitator in the beginning and than slowly ween the program off the professional when ready. Thanks for the input.

Vee / Robbie,

Just to clarify a point here. I agree fully with Vee's statements on RCM and that a structured approach is required.

I fully support the RCM standard, SAE JA1011, and believe that this provides one of the much needed baseline documents that our discipline has been looking for for a while.

It has been my experience that any arguments against this are generally for reasons of commercial interest rather than for reasons of RCM process integrity etcetera.

However, and the point needs to be made, that the standard does not highlight any particular way in which RCM should be implemented. It is not the case that the team approach is the only way to do this, neither is it the way that the only alternative is the sole analyst approach.

I have developed, used, and trained many people in a mid-way approach.

One that uses team members where they are required using fast paced techniques, and uses individual accumen and flexibility where it is needed. (As well as focusing on the end-to-end picture and reducing analysis time by over half in many cases)

This has not compromised the method at all and has allowed many company to embrace RCM where they previously could not. (Because they thought the team facilitated approach was the only way it could be done)

I think that in the 21st century, with all of the means of communication that we have at our fingertips, as well as the advanced number of networking and interactive research tools, that for us to restrict the input to our analysis to just those people who could spare the time to sit around a table (often for days at a time) is not going to achieve the best result for the company.

My thought for the morning (I generally only get one)

Cheers...

Now now Vee - scaremongering again... - appealing to emotion!

quote:

Faced with a Texas City type scenario, if you are defending the case in court, which one would you rather have followed?

Disasters occur through a series of both evident and hidden causes that connect and compound. RCM deals primarily with failure modes on their own with the exception of hidden failures at which point the analysis is done at a second level... nowhere near the depth needed to prevent these disasters. RCM does not analyse sequences of evident failures.
People should not make claims that link RCM to disaster prevention. The hypothesis that RCM is the best way to reduce exposure to disaster is not supported.

Regards
Steve

Steve,

Fair comment.

On review of the disaster from the Houston refinery, as well as the preliminary evidence from the Buncefield explosion (I was near there at the time by the way, scared the living daylights out of me) it does seem that a properly focussed RCM analysis would have probably avoided these.

Does that mean RCM is the only way, nope. Does it mean that it may have avoided these in this case, seems that way.

Cheers,

Ozgipsy:

The team approach is good in some situations. I am not a fan of a full facilitated approach where everyone is locked in a room and singing campfire songs. I think everyone does need to be involved though. I believe in the importance of the team as everyone being involved and not necessarily a facilitated team approach of everyone sitting in the room and hashing out every single tidbit of information. I am also stressing the importance of having someone experiencedrunning the show. Thanks

Steve,

quote:

Now now Vee - scaremongering again... Cool - appealing to emotion!

Nope, not at all; you should try reading the thread on the Texas City explosion, when there was a lot about criminal liability, corporate killing etc.
Any process that can help manage risk can help prevent disasters. Thus HAZOPs, FTA, RBI, IPF, RCA etc can all help avoid major incidents.

quote:

People should not make claims that link RCM to disaster prevention. The hypothesis that RCM is the best way to reduce exposure to disaster is not supported

RCM is just one such risk reduction process. I do claim that RCM will help minimize major events, not that others don't. As to its claim to fame not being supported, pray tell me how you prove an event did not occur as a result of RCM or any other process? The whole idea is to to prevent major events. We all know of the ones that DID happen, but how do you find the ones that NEVER did?
When we have a before and after situation, as in the 1960s with the airline industry, I dont think there is much doubt that RCM worked brilliantly. But looking into the future, I am not sure how one can 'prove' that RCM will do the trick, one can only onfer based on the airline record. If you are a follower of the barrier theory to event escalation (Trevor Kletz' approach) you will agree that the more effective the barriers, the less the risk. RCM provides the means to build such a barrier by identifying the right tasks and their frequency.
You are absolutely correct when you comment about there being evident and hidden causes. But the key is we DONT know the hidden ones till too late, UNLESS we do something proactively. While RCM cannot identify all hidden causes (like how people will behave in a given sitation), it can at least tell us how likely it is that all the safety devices and systems will work on demand.