The concepts of effective planning and scheduling are basically common sense when you get right down to it. Nonetheless, not many us do it right. The common-sense concepts are that “no one is perfect,” “you have to give people enough work,” and “not all new reactive work is going to start today.” Properly applying these concepts gives a wonderful maintenance effect that makes our companies much more profitable.
I learned in my earlier career at the power plant in Florida that maintenance planning and scheduling was basically an excellent opportunity to make a lot of people angry. We were drowning in a sea of reactive work and viewed planning as a nuisance that wasted our labor resource. Honestly we did not see any real benefit. But because upper management declared it to be a “best practice,” we endured at trying. Our planners busily made detailed plans that our skilled workforce did not need, and we let the reactive work bypass planning so as to not delay craft response. We found the planners ended up primarily helping craftspersons on jobs already in progress, finding extra parts or certain bits of helpful information here or there, whether the jobs were planned or not.
No one appreciated the planner role. We had to make people be planners against their will. Anger came from being told to use plans that weren’t that helpful when the reactive plant could probably use the planners back on their tools. In talking with other plants, we found them to be in about the same boat (sinking). How could it be a best practice?
Planning individual work orders was our first area of frustration. Our management had told our craftspersons that they “would get perfect plans and no one would ever have to hunt for parts anymore.” You can’t say that! That’s like saying “the check’s in the mail.” All it did was make craftspersons fuss at planners about any job that had any problem and then insist the planners drop what they were doing to help out. In the past before we had planning, craftspersons had simply applied their own ability, skill, and expertise to solve issues and complete work without help. The extension of “no one is perfect” means also that “no job plan is perfect” and led to a lot craftspersons fussing at planners. No wonder no one wanted to be a planner.
Applying the common-sense notion that “no one is perfect” made all the difference in the world for planning individual jobs. Dr. Edwards Deming, an American in the 1950’s, told American companies that if they admitted they were not perfect, they could get better! But American companies laughed at Dr. Deming and so he sailed to Japan and helped companies such as Toyota become giants in industry. They achieved greatness by acknowledging success, but also continually looking for little improvements here and there in a never ending cycle of feedback and improvement.
In the 1980s, America “discovered” Dr. Deming and we see the basic concept of the Deming Cycle bandied about as PDCA (Plan Do Check Act) and Continuous Improvement. Nonetheless, it is easier to talk about continuous improvement than to apply it, and we simply do not do a good job of it in maintenance. Yet the Deming Cycle solves our maintenance planning frustration. Using the word “planner” sounds like we are going to tell craftspersons how to do their jobs (and also that all the parts and tools will be ready every time). Instead we should admit that no one is perfect.
We should refer to planners as “craft historians” that give head starts by clarifying operator requests and save information to make head starts better over the years. The saving of information is a great help because a lot of craftspersons save scraps of knowledge in their lockers to help them in the future. Wouldn’t it be a great idea if someone saved information for everyone in an organized way to make the information available to everyone? Yes, and that is the one of the primary purposes of having a planner.
One misconception seems to be that planners should always provide perfect plans so craftspersons can complete jobs with perfect precision and consistency. But common sense dictates that no one or plan is perfect. Yes, we want great job plans, especially for new persons and as a reference for experienced persons. But we cannot start there. There simply isn’t time to make perfect plans from scratch for every job even if we could make them anyway. Therefore, the best use of a planner is to give head starts and make plans “better over time,” especially with feedback from the craftpersons themselves.
By not having to be perfect, planners can plan more of the work and help the overall quality of the job plans rise via continuous improvement. We want better plans for more of the work via the Deming Cycle. (BTW, Dr. Deming himself called the Deming Cycle the “Shewhart Cycle” after his mentor Walter Shewhart from the 1930s. This thinking is almost 100 years old and shame on us for not applying it.)
Our second area of frustration was scheduling. When we resolved planning, we did not see our craft productivity improve. We had redirected our planners to start planning more jobs with less detail and redirected our craftspersons to resolve more job problems on their own but reporting feedback. This approach led to more work going through the Deming Cycle where craftspersons received more jobs with head starts and helpful tips from past work. Nonetheless, work order completion rate stayed mostly flat. We were not getting ahead of the backlog.
Applying the common-sense notion that “you have to give people enough work” made all the difference in the world for scheduling. We learned we were a victim of Parkinson’s Law (the amount of work assigned expands to fill the time available) as described by Cyril Parkinson in the 1950s. We had made it easier to complete work, but we did not assign more work. So all the work assigned took as long as it ever did: Parkinson’s Law in practice.
We started experimenting with several scheduling methods. We first tried making detailed daily schedules a week ahead of time. This effort consumed a lot of time, and the real life churn of daily work led to constantly revising them, a frustration. We then started simply giving each crew a full bucket of work for the next week. For 400 hours of labor, we would give a crew 400 hours of work orders as a simple batch not specifying days. This simple focus of work dramatically increased our productivity and we soon drained our backlog even though we never got much above 50% so-called schedule compliance.
I have since seen other plants under-load the schedules below 100% to get better schedule compliance, but their productivity does not jump much. I have also seen plants overload crews beyond 100% for a “stretch goal,” and productivity does not jump much. “You have to give people enough work,” apparently means the right amount of work.
Dr. Peter Drucker, another famous management guru from 1950s and his concept of MBO (Management by Objective), explains some of the difficulty determining the right amount of work to schedule. Dr. Drucker explained that most people do not know the actual purpose of their activities. Never has that been truer than with scheduling. Most people errantly think the objective of scheduling is to complete the schedule; that is, to get high schedule compliance. That is absolutely wrong. The true objective of scheduling is to help us complete more work than we would normally complete. If the purpose was simply to complete the schedule, that would encourage us to schedule less work to get higher schedule compliance and not end up defeating Parkinson’s Law.
Our last area of frustration was with reactive work. Some of the frustration with assigning 100% loaded schedules led to better dealing with reactive work. Most companies let reactive work simply bypass planning because they do not want to hold up crews from pouncing upon urgent conditions. At our plant, a majority of our work was reactive and so did not get any advance help from planners. But common sense again came to our rescue. It was common sense that “not all new reactive work is going to start today.” Does recognizing this notion mean that maybe we can plan and schedule some of the reactive work? In fact it does.
Coupled with the notion that no plan is perfect, we can actually plan some of the reactive work. The trick is to never, ever, ever, ever tell a supervisor to wait. So planners upon seeing a non-emergency but urgent job pop up in the backlog should simply call the craft supervisor and ask if the crew might start the job today. If not, the planners knock out a quick job plan. Planners don’t have to be perfect and often simply clarifying the scope and adding craft and time estimates is a big help. Not only that, often a planner can quickly attach an existing plan from the files with a super amount of learning from past work. Planners can run the Deming Cycle for much of the reactive work without telling supervisors to wait.
Furthermore, the impact on scheduling is equally as profound. We all have some so-called reactive work in our backlogs that is over a week old. (You know you do.) By having that work planned by the end of the week, it can be added to the weekly schedule for the next week and reduces some of the frustration of otherwise breaking a schedule for known work. Don’t break the schedule next week for something you know about this week. Put it in the schedule. You could if you had planned it. Check with the supervisor on a new urgent job. If they are not going to do it today, knock out a job plan. Some of that work will still be un-started at the end of the week. Put it in the next week’s schedule. We now have a more credible schedule that lessens frustration.
In conclusion. Do planning and scheduling right. Easily perfect your programs by applying common sense. Understand that “no one is perfect.” Establish planning with planners acting as craft historians running a Deming Cycle to save information and give better head starts over the years. Understand “you have to give people enough work.” Establish scheduling with schedulers that fully load crews with a batch of work for the next week where it is okay if they do not get it all done. Understand that “not all the reactive work is going to start today.” Don’t let all the reactive work bypass planning just because you don’t want to slow down supervisors and craft response. Have planners check with the supervisor and only plan the work that is not going to start today. Never tell supervisors to wait. And at the end of the week, schedulers can add that un-started, but planned, work to make a more credible schedule for next week.
Plants that can apply common sense in these areas make huge gains in productivity while doing “better” work over the years. A lot of this common-sense thinking has been around forever, but formally proclaimed way back in the 1950s! (Your author was born in 1959.) Don’t be ignorant. Apply well-known common sense. What great opportunities to be a great plant!
This story originally appeared in the April 2022 issue of Plant Services. Subscribe to Plant Services here.