Managers usually use the PM to Corrective Maintenance ratio to evaluate the effectiveness of the PM Program. An effective PM program is demonstrated by an increasing tendency of number of PM hours with decreasing tendency of number of corrective hours.

I work in the pharmaceutical industry, here the operators perform an equipment cleaning between a batch of product and the next batch. Under some circumtances (for one example different product will be the next processed) the scope of the cleaning required is larger (i.e. more time required).

I sure most Maintenance Managers will see adding the operators cleaning hours to the PM hours as cheating the PM / Corrective ratios.

I see a difference between operators cleaning equipment surfaces exposed to process materials vs mechanics cleaning technical areas of the equipment and ensuring no leaks are present.

What do other think?

Darth,

You raise some very good questions. I have some strong views on these, so I apologise in advance for the intensity of my responses.

1. Measuring PM to Corrective ratio by itself is not useful, unless read in conjunction with total maintenance man-hours. When you do RCM, you will find that many failures can be allowed to happen, because their consequences are very low. In such a case, the CM will increase. But that is what the RCM process result is, so how can that be bad? However, if you trend both the PM:CM ratio AND the total maintenance hours, THEN you can draw better conclusions.

Quote >> An effective PM program is demonstrated by an increasing tendency of number of PM hours with decreasing tendency of number of corrective hours ,,Unquote

No. this is only true when very little PM is being done in the first place. A high PM level implies better planning. But a high PM level can also mean excessive downtime. Eventually what matters is that the TOTAL manhours and the TOTAL DOWNTIME are reduced. An RCM analysis will produce the optimum mix of time-based, condition-based, breakdown and detective tasks. There is therefore an optimal PM:CM ratio for each system or equipment derived from the estimate from the RCM study. That should be the target, not some drive to keep on increasing PM. Planned CM is good, unplanned CM is bad. Planned CM is what the RCM study says is the right strategy.

Some Maintenance Managers manage Performance Indicators, not performance itself. The fudge around whether to put cleaning in PM is a way to make the KPI look good; it does not change the performance one bit. This is typical of sites where they bother more about PM:CM ratios than business performance. In my view, it does not matter whether you include it or not; the trend is all that matters when read along with the total manhours(which should trend downwards) and total downtime (ditto). In general, the PM:CM ratios are too low when compared with the (RCM based) ideal ratio. Hence a rising trend will be good, since it will approach the ideal. But recall that the ideal for each system or equipment can vary quite a bit.

There is no Utopia!

Hi Darth,
There are two points about which I would like to make comment here. First is the the PM / CM ratio as a measure of the effectiveness of PM. The discussion needs to consider how you define CM - to you mean corrective maintenance that is planned or unplanned or both. Unplanned is breakdown maintenance.
The second point is to agree with you to say that if the managers are counting cleaning that is driven by changing product, then they are cheating. We would only count cleaning if it were related to the ability of the machine to perform its duty, it is required to perform inspection or the cleaning of itself reduces rates of decay for example corrosion.
The numbers I have provided do not contain cleaning manhours that are driven by change over of product.
Regards
Steve

Vee,
I am with you on this one, but it depends a little on the how CM is defined thus my question.
Regards
Steve

Vee, I second your opinion and the definition of CM is important, and also the if CM is perceived as negative in the organization. In that case people will do anything to look good on paper.

We once had an issue about a "complex" mobile equipment (hydraulics, pumps, engines, travelling blocks, clutches, air etc..) which normally generated 7+ workorders a month.
Someone complained about the amount of CM, and the hard working craft reduced the CM to 1 wo per month.... Until we found the "mamut" workorder called "repairs" where the activities were charged against

Steve, Vee, Svannels:
Focusing on the second point (the first one could drag the discussion to another discussion {which I agree is important})...

In my case equipment cleaning (by operators) has more than one reason to occur making it difficult to judge if I would be cheating if I add those hours to the PM hours. The main driver is operational: to avoid product cross contamination, meet regulatory requirements; but as a by-product the cleaning aids maintenance by:
* being a pre-requisite before the maintenance inspection. Example: a part gets a blow/scratch during operation and requires polishing, (also verify how/why the blow ocurred and fix to prevent next blow). Without cleaning the scratch could get unnoticed.
-----
The operators test emergency stops and do other checks as part of the equipment setup, by doing that may discover hidden failures, then they enter maintenance notifications at CMMS for the applicable repairs.
* Count these operational checks as PM (or detective tasks) since a hidden failure was detected?
* Maintenance time starting before even the maintenance notification or request is created at the CMMS? Not sure.

Steve, I'm interested to know under what circumstances you got your high figures and I hope you can response to my views posted on page 1 of this thread on 16 March 2006.

Another thing, did you say "unplanned CM is breakdown maintenance"? In my organization, we distinguish CM from BM ie CM is carried out while equipment is still running or being stopped by humans in order to do the CM whereas BM is done after the equipment has tripped or stopped by itself.

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

Hullo Darth,

If I may preface my observations by quoting some principles, as follows.
1. Cleaning of equipment is a maintenance activity, irrespective of who does it. The work is invariably assigned to Operators, for several good reasons, which I will not go into at present.
2. Testing for hidden failures is a maintenance activity, again often assigned to Operators.
3.. Lubrication of equipment is a maintenance activity.

All of these 'retain, detect or restore' the functionality of an item, hence maintenance. Producing a quality product requires (internally) clean equipment, so internal cleaning is also a maintenance activity. Please regard maintenance as a process, not an activity. Activities fall into 'departmental boundaries' (silos), while processes are handled by the most suitable persons available.

Hence Operators are often the people who clean, lubricate and test. If we could capture their inputs in the CMMS, we should do so. Often this is too difficult, though with EAMs and a good coding system that is properly followed, there is no reason why this can't be done. When this is possible, we will get a true picture of maintenance inputs. And when this happens we can manage the maintenance process properly.

However, .........
This is not possible in most practical situations. Hence my suggestion to look at trends, not absolutes. The totals matter, and a downward trend is important, whatever be the ratios. Consistency of definitions is another important issue. We should not change the contents of the numerator or denominator in any ratio arbitrarily or often, else trends become meaningless.

The word 'cheating' is perhaps harsh and unfair. Often there is no practical solution to the issue of how to add up all the inputs. If you do have a method, and still whose to allocate cleaning, lubrication and testing effort to operations, that is another matter. If we manipulate the definitions of ratios, that is cheating. Regards.

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

Josh,

I think you are saying the same thing as Steve.

Planned CM is that you decide in advance. So you stop the machine to do it.

Unplanned CM is when the machine decides to stop, not you. That is BM.

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

Vee, to me BM is the extreme case of reactive maintenance ie the equipment has stopped by itself & being non-operational (agree that it's not necssary bad as per RCM). Is run to failure a BM? CM is done while the equipment is still operational and carried out as and when required.

Josh,

From a business point of view, we can take all the emotion away from our decisions. We do what it takes to get the required Uptime at an acceptable cost.

1. Preventive time-based work costs us in two ways
- it stops a machine when it may be in good condition; in this case the downtime was unnecessary
- we may replace parts that still have a lot of useful life remaining, adding to costs.
- we may disturb the internal alignments when we replace parts and cause new failures.

PM can ONLY be justified IF the consequences of failure are unacceptable.

2. Condition-based maintenance(CBM) is based on the machine TELLING you it is sick. It will therefore always be cheaper than PM. The failure has ALREADY started, so the part you replace is known to be damaged.

3. Detective maintenance(DM) identifies failed items that could result in multiple failure. Usually these have high consequences, hence DM is very important.

4. Both CBM and DM can lead to Corrective maintenance (CM). Such CM is important and should be executed promptly, since the item has failed or will do so shortly.

5. If the failure consequences are so low that it is not worth preventing failure, what is wrong with a run-to-failure(RTF) strategy? RTF, when it is a planned activity is the cheapest business decision. The item will break down with RTF,but what is the problem with that? RTF is planned BM, but it is not SCHEDULED BM, as we dont know its timing.

6. In addition to all of these planned activities, there can be UNPLANNED BMs. These are the ones you need to eliminate. They have four main causes
- our work was of poor quality in the first place
- our failure-mode level strategies were wrongly selected
- the execution steps were not correct, either in content or sequence or both
- we did not do the work on time

7. There is nothing wrong with reactive work per-se. Often it makes good business sense. And pro-active work does not mean taking apart perfectly good and reliable machines. These terms should be well understood, else they can lead to mindless strategies leading to poor business decisions. Let us get the emotional hype out of running the business.

Hi Josh,
Sorry for the delay - being doing a bit of travel. Great questions – I will do my best to answer them below. If you have any more, let me have them.

[quote]I guess your figures are high because not all maintenance works such as major repairs, overhauls, shutdown works, major replacements etc are considered. Do your figures include inhouse manhours and external contractors manhours which are normally used for outsourced works?

<My Response> The numbers do not contain major repairs, major replacements unless they are time based. They do include overhauls and shutdown works if the tasks are considered predictive or preventive. If the tasks are corrective, then they are not considered.

[quote] Secondly, maybe your figures are more realistic in the mining & manufacturing scenarios because in these industries, the drivers are the operators cum first line maintainers and heavy repair works are done by central workshop.
<My Response> In most industries heavy repair works are done by central workshops. Industry type does matter as you can see from the numbers / ratios.


[quote] For your process plant scenario, is it a continuous or batch process? What is the level of automation employed?
<My Response> the process plant scenario is a both continuous and batch. There are seven batch processes that feed one process. The line never stops but each batch line does.

[quote] In all the scenario, what is the organization set up? Do they have two separate operations and maintenance departments or maintenance is a subset of operations with a centralised workshop?
<My Response> In all the scenarios, there are separate maintenance and operations departments. Total integration of these departments is starting to happen – I know of one plant that did not separate the two but I do not know how it is working. Many plants put maintenance teams under operations control, but I would not consider this total integration.