Parkinson's Law for Maintenance

Many years ago, when I was principal reliability engineer, a plant manager suggested his plant had too many preventive maintenance (PM) plans. He was having difficulty getting all his PMs done on time.

I thought his statement was odd at the time because it didn't match the verbal coming from his maintenance and reliability practitioners. So, I investigated.

I worked with the maintenance process specialists to simulate all PMs in our maintenance work management system (CMMS). We performed the simulation for six years to ensure we counted all of the orders and work durations. Then we compared the total work durations from those PM plans to the available maintenance workforce. I was trying to answer this question: How much proactive work are we providing our maintenance craftspeople as a percentage of the available workforce capacity?

Fortunately, most proactive routine maintenance plans were in place and in service to be automatically triggered in our CMMS. We used our CMMS rather well compared to many other industry organizations. We added a few manual call PMs not included in the simulation. We excluded planned outage and turnaround work for this exercise. Overall, we felt we had identified pretty much all the proactive routine maintenance we expected to perform. The simulation with additions provided a reasonable estimate of annualized PM work hours (actually, it included both preventive time/usage-based work plans and predictive on-condition work plans).

When we tallied the available hours, we took the number of craftsperson hours (this plant had a maintenance crew of 35 people) and multiplied that by the number of working hours we expected within a typical year. We were careful to subtract suitable amounts for vacation, statutory holidays, illness, training and other administrative duties to converge on the hours we would expect would get charged to maintenance work orders.

The results were surprising. We identified PM work would only keep our craftspeople 13% of their time! That value was much lower than anticipated and significantly below what the plant manager would have believed.

At first, we thought there was some error. Did we get all the work? Yes. Did we miscalculate available work hours? No. Were the planned durations too…