Planning and Scheduling

3 surprising parts of good maintenance planning

Doc Palmer says a willingness to implement unusual concepts leads to best-practice programs.

By Doc Palmer, PE, CMRP, Richard Palmer and Associates

Three unusual concepts make the difference in determining whether a planning program will be successful. Most planning programs are not successful to the point of being frustrating or even counterproductive. It is the fact that these concepts are unusual that gives programs the opportunity to become successful and greatly contribute to plant reliability. These concepts are: (1) pushing planners to plan all of the work, but allowing plans to be imperfect; (2) fully loading weekly schedules, but allowing supervisors to break them; and (3) planning some of the reactive urgent work but never telling supervisors to wait.

The first unusual concept is that allowing planners to be imperfect supports planning nearly all of the work. What makes most maintenance planning programs frustrating is planners and craftspersons alike are given the impression that plans should be perfect and craftspersons should not have any problems when working planned jobs. No one is perfect, and maintenance is certainly not assembly-line work. In frustrating planning programs, planners have to continually apologize that something in the plan was wrong, and they usually end up abandoning planning to help with jobs in progress.

Planners simply cannot plan all of the work if they are trying to make each plan foolproof and they are helping jobs in progress. The best planning programs, by contrast, recognize the Deming Cycle of Plan-Do-Check-Act and concentrate on giving head-starts and improving plans after execution. Craftspersons concentrate on doing the best they can with the head start they’ve been given and suggesting improvements to the plans for next time. In practice, the planners become “craft historians” rather than “perfect plan providers.” In this manner, planners rarely help jobs in progress and therefore can plan nearly all of the work. Planning most of the work allows for job-plan improvements that promote better work quality and support scheduling.

The second unusual concept is that of scheduling all of the available hours for the next week but allowing supervisors to break the schedule. Most planning and scheduling programs are not successful in helping crafts become as productive as they could be, and plants don’t even know it. Most programs tend to schedule only about 70% of the available hours to allow for both break-ins and good schedule compliance. These plants have high schedule compliance, often above 90%, and they feel good about themselves. But the best scheduling programs fully load weekly schedules and expect schedule compliance between 40% and 90% – certainly not over 90%. These successful programs usually find they complete as much as 50% more work each month than would otherwise be the case. Many plants that start 100% scheduling actually run out of backlog within 8 weeks or so. That does not mean they are overstaffed but rather that they have not been very good at generating proactive work.

Proactive work is that work that operators do not scream about and that usually dies in the backlog, if it is written at all. Successful scheduling programs can help plants do more proactive work than a normal plant would in addition to addressing a heavy load of reactive work. The phenomenon that allows fully loaded schedules to produce more work completion is Parkinson’s Law. This law states, “Work expands to fill the time available for its completion.” Successful plants must give their crews enough work.

The third unusual concept is that of planning some of the reactive urgent work but never telling supervisors to wait. Most planning programs bypass planning for all reactive work (work that should be done this week). They fear that jobs will bog down in planning. But because plans do not have to be perfect, the best planning programs try to plan some of the urgent work. Emergency work should be started immediately, but some of the urgent work does not have to start today. Planners can check with supervisors on a continual basis as new urgent work arises:

Planner: “Hey John, I see there are three new Priority-1 work requests. Are you about to start any of them today?”
Supervisor: “We’ve already started the valve job, but we probably won’t start the pump and filter jobs until tomorrow.”
Planner: “OK. Thanks.”

The planner can then take a look at the pump and filter jobs to try to create quick plans. The key is only putting a minimal amount of planner time into the new plan. Simply looking in the plant files or computer might show that a comprehensive plan is already available. Even if one isn’t, the planner might be able to at least clarify the scope of the job. The planner might simply realize that the job may or may not have the parts needed on site.

The planner can also at least identify skills and hours needed to complete the job. In this manner, the planner can run the Deming Cycle and also support scheduling. It would be a shame not to quickly find some plan information today if the supervisor was not going to start the job today anyway. And some of the quickly planned jobs end up not starting tomorrow or even this week. For jobs where there’s a delay of a day or two, the planner might be able to add more detail to a rough plan. Jobs that will not even start this week can be added to next week’s schedule.

It is the unusual nature of these planning and scheduling concepts that provides competitive opportunities for most plants. Plants can improve work quality over time if they can be patient enough to run a Deming Cycle with planning and allow imperfect plans to grow. Plants can defeat Parkinson’s Law to complete a much greater amount of work each month if they schedule more work. And plants can even plan some reactive work without telling supervisors to wait. A willingness to do unusual things leads to best-practice programs with best-practice results – most unusual.