Planning and Scheduling

# Understanding the math of maintenance planning and scheduling

## Doc Palmer explores what adds up to a truly great maintenance planning and scheduling program.

Some interesting math concepts are at work within three areas of successful maintenance planning and scheduling programs. One is that wrench time in companies with good planning and scheduling programs is surprisingly low. Another is that schedule compliance scores in great programs are surprisingly low. A third is that planners’ time estimates for individual jobs often turn out to be highly inaccurate.

First of all, realize that merely good companies encourage competition from other companies that want a share of the industry’s profits. In contrast, a great company discourages competition from other companies because the great company has competitive edges that are often hard to match. Planning and scheduling offers a competitive edge in making a maintenance force more productive to do more proactive work than normal. Even so, proper planning and scheduling has a number of unusual concepts that are difficult to master not because they are complex but because they are not obvious.

One concept is that so-called wrench time, or tool time, at most plants – even good ones – is only 35% for a maintenance force. It is hard to believe that wrench time could be so low, but for most plants, it is. Proper planning and scheduling moves wrench time up to 55%. That is a 57% improvement in productivity (55/35 = 1.57). That means that a maintenance force currently completing 1,000 work orders per month could be completing 1,570 work orders per month! This increase comes from taking advantage of the not-obvious low wrench time for merely good plants.

Another math-involved concept relates to schedule compliance. Proper schedule compliance is between 40% and 90%. A score of, say, 60% means that the schedule has probably been correctly loaded to 100% of the available craft hours to defeat Parkinson’s Law (the amount of work assigned expands to fill the time available). The 60% score means the maintenance force is probably at 55% wrench time and completing around 1,570 work orders per month. But a schedule compliance score of, say, 95% indicates that the schedule has probably been too lightly loaded to defeat Parkinson’s Law. The 95% score means the maintenance force is probably at 35% wrench time and completing only 1,000 work orders per month. This concept of “wanting” a lower schedule-compliance score, though, is fairly counterintuitive.

Another math-involved concept is that of the accuracies of time estimates. Maintenance work simply is not assembly-line work with enough repetitions to allow detailed time and motion studies or allow enough craft experience on the work. So the estimates made by planners have a great variance in accuracy from actual labor hours, perhaps as much as plus or minus 100%! A job planned for five hours might take as little as one or two hours or as long as eight or nine.

Nevertheless, there is very normal statistical distribution, meaning that as many jobs go over as under. A batch of work for a crew of 10 persons for the week might contain 400 total hours of estimated work. The total actual labor spent on those jobs might very well be within 15% of the 400 hours estimated. The implication is that individual job estimates are not that accurate, but they can still be used to create weekly schedules and generally assign and control work. Interestingly, companies that “grade” planners on their planned versus actual scores encourage planners generally to assign too much time to individual time estimates. Giving too much time on job estimates gives a better planned versus actual score, but it allows crafts to spend more time on jobs. Giving too much time on estimates also contributes to underloading the weekly schedule. Underloading the schedule leads to high schedule compliance. But companies with high schedule compliance are probably at 35% wrench time and only completing 1,000 work orders per month. The planning and scheduling has accomplished nothing even though both planner accuracy and schedule compliance look great.

Be aware of these less-than-obvious math concepts of proper planning and scheduling. Mind the store, not the score. Be a great plant; don’t settle for good.