Resource for Rules of Thumb on Equipment Maintenance Cost 3

[AgsMyDude]

Does anyone have a good reference with rules of thumb for estimating maintenance cost requirements of various types of process equipment compared to capital cost? I have a copy of Plant Design and Economics already but I am looking for something more granular based on the type of equipment. I feel like surely something like that exists in a book somewhere, but I have never seen anything good for it.

Thanks in advance!

 

[EmmanuelTop]

One of the best indicators out there is to compare average(d) annual cost of maintenance and capital cost of the equipment. It is frequently titled as RAV, or Replacement Asset Value, and it is expressed as a percentage.

The more data (years) you have, the better, because having just a year or two of data can overemphasize or underestimate actual/real averaged annual cost, compared to the case when all (at least preventive) maintenance activities that would be performed on a piece of equipment will eventually happen over the span of a couple of years, thus making the average a real figure.

Coming to the RAV KPI value - it really depends on what is "typical" or "expected" lifetime of a given piece of equipment in a given operating context/environment. You should be looking at a case if the design was done properly and the equipment is operated properly (because that is how it is supposed to be done), and then see what would be an expected lifetime of equipment under such circumstances.

As a rule of thumb, if RAV > 5%, it means that maintenance of the equipment would "consume" the asset value in less than 20 years, and this could serve as an indicator if the equipment is perhaps overmaintained, or it suffers due to defects in design in operation, therefore eroding the asset value faster than anticipated.



There is no such thing as "common sense" - Apollo RCA

 

[1503-44]

That could also relate to design quality in the same manner, as long as a relationship between design quality and design life exist. In many cases all three are related, design quality, directly related to useful life and inversely related to maintenance cost. Higher than expected maintenance may reflect more directly on design quality, rather than poor maintenance. Having an idea of expected maintenance cost could help determine if issues are caused by lack of design quality, or a poor maintenance program. When selecting between design options, I use 10% of capital cost per year as my maintenance cost. Higher than it would be at 5%, to give a slightly greater weight to increasing design quality into one time capital cost, as they are more easily controlled at design time, rather than increasing maintenance expenses for the next 20yrs, which are harder to estimate how they might vary over the equipment's entire lifetime.

But rather than use a fixed 5% or 10% per year figure for all equipment, wouldn't it be better to use a factor of 1/equipment useful lifetime. That way, if the life is estimated to be 10yrs, the factor would be 10%/yr, or if life is estimated to be 20yrs, then it would be 1/20 => 5%, or if life was 30 yrs, then 1/30, a factor of 3.33%. Making the factor correspond to expected design life ties it more tightly to the design quality decission. It would also serve for those cases where it is not possible to expect certain devices to last, or be useful for 20yrs, especially those items that are subject to premature retirement due to changing technology at 5yrs or so, rather than actual survival time.

The 10% factor also fits the design choice decission, since it is often easy to get 10% more capacity at almost the same capital cost, or simply by optimizing "fudge factors", rather than trying to increase capacity by any amount after construction

So if you don't have the actual maintenance cost data, relating maintenance expense to design lifetime might be an indirect way to get better granularity without having such a giant maintenance database.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[geesaman.d]

In the type of plant machinery where I specialize (agitators) the cost of maintenance is widely variable based on the quality of the original design and the suitability of the machine for its actual service. I regularly see huge mismatches between what equipment was designed to do vs. how it's being currently used, and the maintenance costs can explode when that is not aligned. Also the maintenance practices of the plant have huge effects - if the maintenance budget is hawked without thought, it will quickly devolve into a reactive-only maintenance team that spends their money replacing heavily damaged equipment as quickly/cheaply as possible to restore production. Those repairs will barely hold together for the next scheduled outage, much less reach sensible estimates for equipment life and cost. There will not be resources for proactive maintenance, which is somewhat predictable.

I deleted a couple of paragraphs that rant about the sources of this. Incompetence, arrogance, bureaucracy, private equity juicing, reduced corporate engineering knowledge and experience, executive cutbacks-for-immediate-bonuses... there are just too many. But I would take great care to consider whether your maintenance cost model is built for theoretically sound proactive maintenance of appropriate equipment or the cash-starved, highly-maintenance-intensive business mode. It seems that far more companies exist in the latter than the former.

 

[1503-44]

The best way to estimate this without data is to use the maintenance list and schedule. Major equipment usually comes with a list of routine maintenance requirements and part replacement schedules. Your expected maintenance costs should correspond to that list and schedule.

If maintenance costs do not appear to correspond to the maintenance schedule, something is amiss. That's the clue to investigate why, usually allowing the fault to be discovered.

High maintenance costs,
Is the equipment built poorly, causing premature failure.
Is the equipment being operated within design capacity range, or outside the hi/low range.
Is the equipment sized properly, or is it being operated outside limits.
Is maintenance being delayed until failures occur.
Low maintenance costs
Is the equipment not running at design capacity.
Is maintenance complying with the maintenance schedule, or postponing regular maintenance.

Costs must not be too high, or too low, so one way or another it should always return you back to the schedule of maintenance and repair activities to set up the baseline, or throw up those error flags, but those are for the maintenance depth to solve. You have your cost base defined by the schedule.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[AgsMyDude]

Thanks for the responses all, unfortunately we do not have maintenance requirements or data for existing equipment. This is a conceptual life-cycle cost estimate activity. Likely the best approach will be to just apply a uniform percentage of ~5% across all equipment types, similar to the rule of thumb Emmanual mentioned.

 

[1503-44]

Definitely quicker. Never calculate to a level of detail that you don't need.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."