When is the right time for a PM Optimization project?

How about when you think you are doing excessive works?

Our review current strategy every year to optimization PM. why we do this?

To optimization PM, you must to do failure analysis first, I correct?

Wheres Steve Turner...?

Darth,
We will start with the premise that you have selected the right equipment and configuration, and commissioned it properly. Let us also take the situation where the equipment is operated within its design envelope. For these items.
1. The first order of business is to determine the right work to do and the right time to do it. For this purpose, prcesses such as RCM, RBI, IPF and FMECA can be used.
2. The next step is to ensure we do the work to the right quality and on time. Once we do all this we should get the equipment to perform as well as the designer intended it to perform.

In theory there is no need to do any further adjustments to strategy, so PMO is not required. In practice, sevral things may happen over time, such as.
3. The operating context may change.
4. Our starting assumptions may prove to be incorrect in practice.
5. There are errors in our original analysis; the evidence for this is lower than expected achieved reliability. Trends in corrective maintenance, costs etc will tell us when this happens.
6. We are reaching end of equipment or plant life, so our expectated performannce levels may change.

It is not economical to do full blown RCM or similar work to resolve such situations. A PMO is far more effective.
Secondly, if you are unable to do an RCM right at the atart, which is of course the best time to do it, a PMO type of exercise is better than doing nothing.

If the PM Program contains monthly, quarterly, annual and every two-years maintenance tasks, should the PMO be done at least two years after the PM Program start?

Darth,
What drives a PMO review is not your current PM program. The main things you should think about are:
1. Has the operating context changed since you set up the PM program?
2. Has there been any significant change to the equipment itself, in terms of upgrades, set points, configuration etc.?
3. Is the reliability performance as expected or trending downwards?
4. Is any part or all of the equipment becoming obsolescent?
5. How far are we from retiring the equipment?
You can judge whether a PMO review is justified once you have answers to these questions.

quote:

Originally posted by Eugene:
When is the right time for a PM Optimization project?

Hi all,
A major problem with almost all companies that use physical assets to derive revenue is that their PM programs are badly focussed. The problems emerge before the equipment is commissioned when the maintainers are not involved sufficiently in the design. These problems grow as the project budget is cut and design capability is reduced, as the commissioning is done in a rush or not at all, as noone has the time to do proper coordinated maintenance analysis, and when this analysis is done in a rush with no consistent method and no coordination between craft groups and operations, and where attitudes are conservative and risk averse. And then of course the excessive program grows because of knee jerk reactions to burn in failures.
Our evidence based on PMO2000™ analysis of thousands of equipment items at well over 200 sites is that far less than 50% of the existing maintenance activities survive a review using RCM task selection criteria. Commonly 10 to 20% of tasks that should be done, are missing from the schedule. Some maintenance is overly intensive and some is not intensive enough. It is not uncommon for organisations to be doing twice the preventive and predictive maintenance activities that they should be doing.
My suggestion is that if your equipment strategies have been put together in an adhoc manner (or you have simply applied vender recommendations) you should be doing a formal PMO as soon as you can.
My suggestion is that it should be kept up to date by having zero tolerance to unexpected failure. That is, when failure occurs, the PM’s are reviewed formally. This is a simple process if the right tools are being used and the organisation has a Quadrant 2 structure (that is it is set up for planned work rather than breakdown maintenance)
I suggest that six months after initial analysis, a formal review should be conducted and then annually or every two years thereafter unless of course, there are modifications to the process or the equipment as Vee has pointed out.
Hope this helps. There is lots of info on our website for those who are interested.
PM Optimisation Website

Steve, I understand from other post, you worked for a RCM company before setting up your own OMCS (if I remember correctly). So out of over 200 sites you mentioned above, how many of them have their maintenance program RCM done by you and how many afterwards ask you again to review them using PMO? Any substantial differences?

Yes I used to deliver RCM programs.
No, I have not gone back to a company and reviewed an RCM program that I previously facilitated. Sorry - but I am glad because it would appear the RCM work I facilitated was considered to be ok probably.
However, we are currently involved in a PMO review of an RCM analysis done five years ago for a major oil and gas company. In this case the RCM has never been implemented because the recommended maintenance program delivered by the RCM people was poorly done. The previous analysts seemed to have got the analysis all wrong. Now this is not the fault of RCM.
Leaving that point aside, it gets back to Vees point. Philosophically speaking, if RCM has been done on an asset, PMO would only be needed for reasons Vee has stated....together with other reasons such as moving to more operator based maintenance or adopting new technology such as oil sampling.
Now the great caveat here is that the original RCM may not be RCM. There are a few programs out there that call themselves RCM but have decision logic that does not conform to the standard. Ones that use cost minimisation algorithms should be carefully examined for example.
Back to your quesiton, if I were to review an RCM study I facilitated some years ago, I would expect to find the group wanting to make some changes but not a lot. This would be due to personal experiences more than likely. BUT, if the site has been keeping the analysis up to date, that means reviewing the program when failures occur, it should be fine.... with the exceptions of the changes Vee has identified... Therefore, because we are all professionals, we would never leave the PMO review to be an annual event would we? We would do the PMO review when the plant operating context changed or when the mod was introduced wouldn't we.
Interesting discussion Josh and Eugene.
Regards
Steve

Have you reviewed the effectiveness of PMO results for the over 200 sites? Are these sites able to sustain their reliability and availability over the long term say 10 years?

What are the hurdles for PMO implementtion?

This message has been edited. Last edited by: Josh,

quote:

Have you reviewed the effectiveness of PMO results for the over 200 sites?

Josh, we have only formally reviewed the effectiveness of some. We can only do this if we are on a long term conract or asked. We often only get asked when the early results are not as expected.
Reasons for not getting the required effectiveness are usually one or more of the following which are listed in rough order of importance:
1. Not getting the operations sufficiently involved and not having formal operator maintenance tasks. Operators in some process industries do 70% of the total PM workload. There are only a few industries where the number is less than 50%.
2. Procrastination on approval so the work sits on the shelf for too long.
3. Not collecting plant performance data in a practical manner and not following through with failure reviews when breakdowns occur.
4. Poor communication - now this is not usually one of our failures because we make sure this is done as well as possible - in some cases however, the client says they will take care of this and because they don't there needs to be a recovery plan. This could be one of the most critical elements and be No 1.

I can say that the effectiveness rate of PMO is significantly higher than the RCM effectiveness. With RCM the company I previously worked for was getting one in five continuing on and few of these went past the top three items of equipment. With PMO2000 we have only one experience where the client has not progressed passed the first pilot.
Most of our pilot programs these days end up in corporate roll outs. Companies often use the same equipment across the globe. It makes a lot of sense to roll out something like PMO rather than each site doing their own RCM.
The trick is to have software that can handle corporate knowledge across the globe. We have that.

quote:

Are these sites able to sustain their reliability and availability over the long term say 10 years?

Our company has just turned 10 years of age this year and PMO has only been seen as a credible approach since around 1996. So 10 years is a bit too long to try and make some assessment.
However, I do know that not all sites sustain their initiatives. One grand plan for example was scuttled when the corporation was bought out and the plan for that company [my assessment] was to grab the brands and sell them off. The HQ people that were involved in PMO were all retrenched and the site staff numbers reduced to the point where they could barely keep the plants working.
One other case - I shudder sometimes - PMO was live and well at one site after two years until the company decided to rationalise management of its production units. The PMO site management were not kept and managers from other sites who took over seemed to be of the opinion that what ever existed in the site that lost its managers was no good and needed the alt.ctrl.del . PMO had a lot of support and it managed to survive in name - however in practice, the new management headed down a path of criticality assesment deciding this was the way to go. Powerless, the PMO people left or went with the flow. Fortunately, the criticality assessment was stopped after nearly two years and now PMO is being used in this company as a corporate standard. There would be around six full time corporate people working in this area.
So bottom line Josh, like most initiatives, they suffer from external energy sources that can destabilise them. I have to make the point that PMO is fast and effective. Done our way it gets the same maintenance program as SAE Compliant RCM. If companies are going to compare approaches, they should not underestimate the value in a process that takes less time and gives the same (or better) result. I say better because the shorter time gets buy in from shop floor more easily than the longer winded approaches.
Thanks for your interest Josh.
As you can see, I am always happy to talk about PMO. Forums like these are excellent vehicles for sharing perspectives on things.
My perspective is not always agreed, but I am always willing to discuss.
Steve

quote:

Fortunately, the criticality assessment was stopped after nearly two years

I am curious about criticality assessments are they viewed the same as prioritization assessments? I tend to think they are related, but this is from my perspective.

I am also curious if and why they should be avoided?

Again, this is from my perspective and I think I have a case study that supports performing a criticality assessment. But I am not 100%, perhaps not even 50%, certain how the PMO process works - perhaps my thoughts of a criticality/prioritization assessment are already apart of PMO. I have only seen a small snippet of PMO at the 2006 RCM conference in Las Vegas.

My case study or studies are the observations that the "big" catastrophes at sites I have worked at and work at are not typically the result of equipment that is already being PM'd nor equipment that prior to the failure had significant maintenance expenses or throughput losses attributed to them. These big catastrophes create the large losses in production and large maintenance costs – the production loss usually far out weighs any maintenance cost. I have seen maintenance costs in the neighbourhood of 5 to 10 Million where normal maintenance budgets are in the order of 10 to 60 Million respectively (I can not estimate the cost of the throughput loss, but certainly 5 X more would certainly not be outrageous). These catastrophes tend to be thought of as non-routine expenditures, are amortized, and in the next year or five a completely different one occurs that is said to be non-routine because it is a completely different failure from any historical failure.

One particular case study is a Treatment system that is used to process waste product generated by the plant. Through a criticality "type" analysis we discovered we were missing a PM on a seemingly insignificant drive chain, gear box and sprockets. If we did not add a PM to these minor components we could end up with over 5 Million dollars in lost production within the next 10 to 20 years. We have never spent money on these components, and luckily they have not failed in the past, but if we had not realized we needed at minimum an inspection for the failure of these components on a shift by shift basis and have the spare parts in stock we would have ended up with this lost production. The cost of the inspection is negligible (as the operators already walk past the equipment we only added for them to make sure the drive equipment is still operating) and holding cost of the spare parts are no more than $1,000/year. The system consequence (5 Million dollar production outage) occurs if the failed condition is not found and corrected within the next 48 to 72 hours.

In the future we wish to monitor the drive motor for a drop in amperage which signifies a drive failure has occurred. This plant change is expected to cost no more than $4,000. The failure of the drive chain, sprockets, or gearbox is considered to be highly likely in the next 10 to 20 years (shorter than the expected life of the plant).

Is there an analysis step in the PMO process that looks for opportunities that do not already have costs or a historical PM associated with them? I looked at the website and could not clearly discern this – it seemed to mostly focus on the existing PM and failure data.

quote:

“… is specifically designed for reviewing maintenance programs and failure history for equipment”

quote:

am curious about criticality assessments are they viewed the same as prioritization assessments? I tend to think they are related, but this is from my perspective.

I am also curious if and why they should be avoided?

Hi Jaz,
These are my opinions only
Criticality assessments are generally worthwhile. However before embarking on one, I would think that the reason for such needs to be identified before hand. Otherwise they can get out of control.
If the reason is to rank the equipment in order of priority for RCM / PMO review then do it, but don’t spend more than one day doing it. Most maintenance / operations people can list the important equipment in less than one hour. I have seen some organisations spend two months working the list for review when in that time, they could have analysed 50% of the critical items.
If you are listing criticality to determine corrective work priorities, then this should be done with a bit more time and care.
Listing critically to determine PM priorities does not make sense because if PM is cost effective, then it all should be done.
Listing criticality to determine spares is worthwhile, but criticality is only part of the spares equation.
I am certainly not against criticality assessments but I do think this process is over stated as a tool for deciding PM strategies.
Regards
Steve

quote:

Again, this is from my perspective and I think I have a case study that supports performing a criticality assessment. But I am not 100%, perhaps not even 50%, certain how the PMO process works - perhaps my thoughts of a criticality/prioritization assessment are already apart of PMO. I have only seen a small snippet of PMO at the 2006 RCM conference in Las Vegas.

Jaz,
This was probably the presentation I did when the fire alarm test started 25 minutes into the presentation and everyone had to leave.
The PMO process is no different to RCM in that criticality is assessed when failure consequence are listed. Assessing criticality at failure mode level makes more sense to me than doing it at equipment level.
There are many RCM processes that use criticality to streamline the process and I think this is poor and only done because RCM takes a long time to do and companies want to cut to the important things quickly. If RCM were not so time consuming, far less analysts would use that filter I would think.
Having said that, there are a few RCM approaches that seem to use criticality as a driver of maintenance frequency. I don't agree with this approach in most cases.
Steve

Steve,

Thank you for the comments. I believe I am in agreement with the comments. For example my thoughts about a criticality assessment are not magnitudes off from yours. I was thinking 2 days tops.

I see the criticality assessment, historical data, and equipment with PMs providing the "initial" listing of equipment to be analyzed.

The outcome of the "reliability" analysis would then provide a "final" criticality listing. For example, which equipment requires spare parts or overtime for corrective maintenance (spared or not) - based on a systems failure probability and consequence. I believe this can only be discovered after some reliability and availability analysis. If you haven't analyzed your key equipment and applied an appropriate PM program where possible then every individual on the site will have a different sense of what spare parts should be stocked, what failures are priorities to correct first (to prevent a system consequence), and equipment across the site will have varying degrees of preventative maintenance actions - some over PM'ed, and as the case study showed some under PM'ed.

Yes, I think I recall the fire alarm confusion. But since I wasn't presenting my memory of the event is not going to be as clear.

quote:

Is there an analysis step in the PMO process that looks for opportunities that do not already have costs or a historical PM associated with them? I looked at the website and could not clearly discern this – it seemed to mostly focus on the existing PM and failure data.

In some PMO processes, the aim is simply to review what is currently being done. These are fine if that is all that the company wants to do. The US Nuclear Power industry seemed to have this objective in the mid 90's.
If the objective is to create a comprehensive but valid PM program for the equipment, then it is important to add the missing bits (Failure modes) to the list of failure moded derived from the current PM.
This is done in PMO2000 (the process we developed) firstly by failure history review (normally by working with people who have lots of experience on the equipment because documented failure history is scarce) and secondly by reviewing technical documentation such as P&ID's or circuit drawings. These are the ways RCM tackles the task of listing failure modes so if all else is the same, PMO will not generate a different result than RCM. PMO2000 process certainly seeks to list all failure modes are likely and should have PM. I am very comfortable with the statement that PMO2000 generates the same maintenance program for any physical asset that would be generated by an SAE JA 1011 compliant RCM program.
Many other RCM experts say the same thing when they have done a proper PMO2000 analysis.
The attached paper explains PMO2000 and discusses the above statements in more detail.
Hope this helps.
PS - there are plenty of papers on our website that relate to these topis
Papers on PMO2000
Hope this helps to explain PMO.
Steve

ComparingRCMandPMO2000.pdf (760 Kb, 9 downloads) When to use RCM and when to use PMO2000 - a comparison of methods