When establishing an effective maintenance program, one must determine not only which Preventive Maintenance (PM) routines to accomplish, but how often should they be done. The answer to this question would seem on the surface to be quite simple and, in fact, one proven theory is that the PM to Corrective Maintenance (CM) work order ratio should be about 6 to 1. This theory assumes that the PM inspections should reveal some type of corrective work that should be completed on an asset on average every 6 times it is accomplished. The assumption is that, if the ratio is greater than 6:1 you are performing the PM too often; if the ratio is less then 6:1, you are not performing it often enough. (The “6 to 1 Rule”, proven by John Day, Jr., Manager of Engineering and Maintenance at Alumax of South Carolina, during the period when Alumax of South Carolina was certified as the first “World-Class” maintenance organization) You might accept this theory, put it in place in your maintenance program, and forget about reading the remainder of this article. Or, come along and we will attempt to prove or disprove this theory.
Preventive Maintenance is that activity performed in some routine or regularly scheduled fashion designed to keep equipment in an existing state, prevent deterioration or failure, and identify work of a corrective nature to keep equipment from causing non-productive time in any capacity. This is the detection phase of the PM investment; the conditions we identify and correct prior to failure is the return for this investment. Each PM that we develop and implement in our maintenance organization will require some definite period of time for a maintenance or operations person to accomplish. How many PMs, how often, can we accomplish with our work force, reserving a certain percentage of every day for Emergencies, Unplanned work, and Planned corrective work? Should we attempt to implement some sort of control over how often we do PMs? You may realize as we continue on this path that the frequency assigned to many PMs has as much to do with effective manpower utilization as it does with discovery of potential asset problems.
All too often, I visit client sites where the maintenance department is overwhelmed by the number of PMs required on a daily and weekly basis. It is not that these clients have so many PMs that all cannot be completed. It is that there are so many PMs to accomplish, there is little time allowed for Emergency work and NO time left for solid corrective work to prevent emergencies (that capitalizes on the payback for our investment), or perform other unplanned work. We must also consider the potential to “PM the equipment to DEATH!”, actually creating more problems by performing PMs more often than we should. The obvious question is “How effective is your PM program?” The short answer is – If your PM program isn’t finding problems, it isn’t effective.
OK, back to the point. How often should we perform any one PM procedure? It occurs to me that the frequency of performing a PM should be based on asset failure rate (Mean Time Between Failure) rather than the NON-failure rate. If I run a particular piece of equipment to failure, fix it, then run it to failure again, what is the MTBF? Knowing the MTBF, I should be able to calculate a realistic time frame where, if I perform some routine checks and preventive measures on this asset, I have the opportunity to identify potential problems and fix them, significantly extending the MTBF. Is this not our primary goal in maintenance? More does not necessarily equal better in preventive maintenance.
Let’s consider the Air Compressor PM that is scheduled for accomplishment once every month. It requires that you observe the compressor in normal operation and observe for any obvious signs of potential problems, such as air leaks, knocking in cylinder areas, and proper operation of the automatic drain on the accumulator. It also requires that you shut down the compressor and check the condition of the air filter and oil level. The PM is estimated to take 1 hour to complete. If you accomplish this PM as scheduled for 6 months and do not note any problems, or do not have to change the air filter, or replenish the oil, what can you say about this PM? First, you have used at least 5 manhours of valuable craft labor performing a PM that you did not need to do after the first time, not to mention the 5 hours that the compressor was “off-line” for accomplishing the PM. What other work could you have completed with those 5 manhours? Secondly, you can make a logical assumption that once per month is simply too often to schedule this PM. What to do? Should you change the frequency of the PM to Annual right away? If you did not generate a CM work order in 6 months, perhaps 3 or 4 months is a good place to start. In this case, extending the length of time between PMs by a factor of 12 (to Annual) might result in missing an indicator of a potential problem and prove more costly in the long run. Work into the appropriate frequency carefully. If by extending the periodicity of the compressor PM to 3 months, you identify corrective work every 5th or 6th time you accomplish the PM, then you are in the right neighborhood for an effective PM program for the compressor. If not, keep adjusting the frequency until it gives you the outcome you need for an effective, proactive PM program. Good PM work order history will point exactly to the right frequency as you continue to experiment with periodicity for the PM.
When first implementing a proactive Preventive Maintenance program, you should establish frequency of PMs with a conservative view. Many sites implement the PM program using Manufacturer-suggested PM frequencies that, although usually on the conservative side, are a good starting point. Adjusting the frequency of PMs as good history is developed is a benchmarking exercise driven by the desire to identify deficiencies and correct them before they become emergencies while striving for the best possible use of your maintenance resource hours available.
Let’s consider the types of maintenance actually being performed at your facility. There is Preventive, Predictive, Condition Monitoring, Corrective, and Emergency work in our basket, with every type of facility and industry expending certain percentages of maintenance resources on each type. We have already discussed Preventive Maintenance to some degree. Predictive Maintenance (PdM) technology uses some proven testing method such as thermography, tribology, or ultrasonics to trend equipment performance and “predict” when certain preventive maintenance activity should be performed, thereby heading off a potential failure. Condition Monitoring is the practice of closely monitoring equipment on a continuous basis to provide early detection of symptoms that could cause problems or failure, then performing some corrective actions to preclude the problem or failure. Corrective Maintenance (CM) is the act of performing some repair or adjustment for a condition that was identified during the accomplishment of a PM or PdM evolution, and cannot reasonably be corrected within the allowed labor time for accomplishing the PM or PdM. Emergency work requires little definition; it is work performed in direct response to a failure that causes process down time or imminent hazard to assets or personnel. This is the appropriate time to note that, with the exception of Emergency work, all of the other types of work are predicated on finding and fixing the problem before it becomes a “down time” event. Perhaps we could categorize all maintenance except Emergency under an umbrella of “Preemptive” Maintenance.”
We have taken a “long, strange trip” around the Maintenance horn. Are we any closer todetermining the proper ratio of PMs to CMs in a Proactive Maintenance Organization?
Two true statements that can be found in nearly every maintenance-related publication on the market are that:
- You must evaluate the ratio of Preventive Maintenance actions to Corrective Maintenance actions to determine effectiveness of your PM program and,
- In a Proactive Maintenance environment, PM activities should account for approximately 30% of total maintenance resource time.
The real answer could be that the ratio of PM to CM is dependent on several variables:
- Asset Criticality – If the asset fails, what is the impact on production or safety
- Asset Age – Equipment histories will prove that most failures occur during infancy (newly installed or overhauled) and old-age (self-explanatory)
- Asset History – How many times has the asset failed in the past (MTBF) (Answer the What, How, Why, When, and How Much)
- Asset Technology – Do you need to PM a state-of-the-art digital measuring device as often as the analog one installed next to it?
- Trust – How much do you trust the asset to perform as designed when scheduled to run?
- What percentage of your total maintenance resources is expended on PMs? CMs? Emergencies?
- Can you change PM frequencies within the guidelines you have established for yourself? (ref: ISO, CFR, MILSTD, etc.)
If you work through all of these exercises, you will find that the proper ratio for the majority of your assets will likely be close to the 6:1 ratio mentioned earlier. This is not to say that ALL PMs should be targeted for that ratio. The “6 to 1 Rule” worked for John Day and Alumax, and it works for many other facilities that have chosen Reliability Excellence and the proven processes that come with dedication to a proactive approach to maintenance. Each asset in your facility must stand on its own merits and the proper ratio of PM to CM determined as a direct result of analyzing past performance and PM work order history, not a guess on the part of the Planner, Supervisor, or Maintenance Manager.
© Life Cycle Engineering, Inc.