Dear All,

I would like to ask your opinion and your point of view if failures can really be eliminated through reliability and continuous improvement efforts ?

TPM claims it will aim for zero breakdown, is this really possible ?

With all the maintenance training, tools of the trade, discussions here and there in this forum, reliability methods used, RCFA/RCA technology, is there a way really for maintenance to eliminate the failure.

I'll provide my remarks later, once I read your points of view.

My Warm Regards,

This message has been edited. Last edited by: Rolly12,

Rolly,

I think that some can be eliminated, others can be managed, and some will still come out of the blue to get you.

So, the short answer from my experience is no.

TPM aim to Zero is basically keep the goals and efforts high to get as far as possible.

I think that the company as a whole can eliminate zero breakdown with unacceptable consequences.

Hi Rolly, I think that with the right tools and techniques and a workforce with the right attitudes and motivations a high level of OEE can be attained - though this would come at a price.
Maintenance done right is going to cost a company in the short term - in some cases businesses (especially the small and medium sized ones) may find this cost unbearable.
However the larger multi- nationals I think can attain OEEs of around 80 - 90% which I think is world class level.

Mike.

Rolly,
Should they be? Does it not matter what the consequences of failure are? Gold-plating equipment may work, but is that worth-while?
Human failures account for a large proportion (some claim it is of all failures), so yes, in theory most if not all failures can be eliminated. But reality steps in to challenge our economic credentials.
In my view, managing degradation is better than eliminating failures.

We love to focus on equipment and we ingore processes and people.

Nuke industry report on unreliability:

38% caused by people

36% caused by process

28% caused by equipment

(Disclaimer- this is from my memory which is subject to being a "wee bit off" so I will search for the actual reference to post later)

Rolly - spend at least 74% of your time improving people and processes if you would like to eliminate failures. (at least to a 28% level).

Terry O

And if machines could talk Dutch or English then that 28 % would drop significantly.

Dear All,

Let me rephrase my question, I would like to hear your opinion on which one would you think is correct. I plan this to be my next newsletter, and definetly, I'll give credit to all of you.

So which one of this statement is correct ?

1st : We can definetly eliminate failures by analyzing them through RCFA and RCA .

2nd : We cannot eliminate the likelihood of a failure, but rather we can only prevent or predict the failure from occuring.

3rd : Failures cannot be eliminated and the best that maintenance can do is to reduce them.

4rth : Failures cannot be eliminated the best that maintenance can do is to prolong or delay the process of failure itself.

My Warm Regards,


Rolly Angeles

Rolly - try an opinion poll. It may be interesting.

Mike.

Dear All,

I would like to post my answer now, but I think I will give it some more time, once I post my point, i know it will post some questions but I just want to tell what the truth is with respect to equipment failures on whether we can eliminate them or not.

Hope you can state your point of view.

I always believe that for every single failure there is always a human intervention, which means that human are in a way connected on why our equipment fails, around 99%. Not only from how we maintain, but from the design stage, commissioning stage, debugging stage, how our operators operate, how we make decisions on our equipment, and of course on how we maintain our equipment.

There is a very strong connection between failure and people directly or indirectly connected with our equipment.

Now going back to the question at hand, hope you contribute your point of view here !!!

My Warm Regards,

Rolly,
I believe that it's your No4. All machinery designs are compromises. Materials, manufacturing, build quality etc. Then on maintenence, spares, drawings, past history and again build quality.
We're progressing all the time, Thats life. Zero failures will always be an aim, but it must be about balance of effort.
Good Luck
Joe Mc Cormack

quote:

TPM claims it will aim for zero breakdown, is this really possible ?

Yes, if all work together. It takes time to get to that position. In one plant where I've been doing PdM for 17 years we have zero break-downs unless lightning or something to that effect strikes. When I started machines typically ran 1-3 years; now, 7-9 with planned scheduled maintenance = zero overtime. I was teaching my alignment seminar in Macon, GA at Macon Technical Institute and 2 of my clients were on-hand speaking up and making notice of that fact. I was shocked.

Absolutely not. You will never be able to prevent all failures and therefore never be able to eliminate all breakdowns. Not every failure is predictable and the cost to try to prevent every single breakdown would get so ridiculously high that it would not be cost effective. I agree that you can get an extremely high reliability out of your process but not perfect.

How about reducing risk to ALARP (as low as reasonably practicable, to do business)

I think RCA is very important. I don't like having to fix the same thing over and over again. Granted there has to be some wisdom behind it - spending a week to figure out why a $2 light bulb blows every three months is probably not a good expenditure of resources.
Yesterday I was finally able to track down the source of an intermittent problem that we have been having for some time. The problem used to happen perhaps twice a week, with a small failure in a machine that would cause perhaps 15 minutes downtime each time. Since the failures were so rare, and the description of what was actually happening was so varied (every operator had their own idea as to what was going on) it was not worth my time to sit there and watch the machine to try to sort it out.
Finally yesterday the problem got severe enough that it was happening reasonably reliably, and I was able to find it. It took me about 3 hours to track it down, which is a lot longer than the 15 minutes it would have taken to get the line running again, but now I am confident that this problem will not happen again now that I have added this inspection/cleaning step to the yearly PM for this machine.
Luckily my boss is a Big Fan of RCA, so he doesn't mind me spending time and resources to find root causes.

So now that I have found this particular root cause, that will be, compared to historical values, two less downtimes each week.

I love finding root causes... it makes me feel good.

So my opinion is that RCA most definitely helps to increase reliability, but it's not always worth the time and effort. As long as a bit of wisdom is used in its application, then it is an incredibly valuable tool.

quote:

Can equipment failures really be eliminated ?

Absolutely yes, two years ago the department of maintenance designated me as team leader of a multidisciplinary group for RCM analysis to generate plans of maintenance for a complicated equipment.
This included the selection of the participants from the differents departments of the plant.
The results were to eliminate 37 failures for month to zero failures, how?
Simply doing RCM analysis and TPM, then the uptime increased and the line productivity increased 6% ...and of course... the paycheck of all plant too.

Cristian.

Cristian,

If you applied RCM then you will find that in many cases you are now managing the failures, not eliminating them altogether.

The end result is that you have eliminated all unplanned failures.

My understanding (such as it is) is that the whole point of RCM is to identify those failures that can't be efficiently prevented, and just let them happen. The net effect is to save money, particularly since many failure modes have an 'infant mortality' aspect to them such that replacing a functional part can _increase_ the chance that it will fail. If my understanding is correct and this is indeed the case then RCM can _not_ be said to eliminate failures, it just saves you money in the long run and hopefully reduces failures.
As for TPM... well, no matter how many times you check and replenish the pressure in the tires of your car, and re-balance the wheels, if you drive over a nail you're going to get a flat tire. No amount of maintenance will prevent that.

quote:

Originally posted by Ozgipsy:
Cristian,

If you applied RCM then you will find that in many cases you are now managing the failures, not eliminating them altogether.

The end result is that you have eliminated all unplanned failures.

So what you are saying is that if I know there is some potential (random possibility) a bird might strike a plane and the effect is that it will crash, the fact that I knew it might happen makes it a planned failure? Completely random failure cannot be planned. You can try to manage all the failures in the world but you will still experience unplanned failures from time to time. Calling the dog a cat will not change what it really is.