How credible is your priority system?

What needs to get done when? Doc Palmer helps to ensure that everyone's on the same page.

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

The second principle of scheduling is that a plant must have a credible priority system for work orders. The priority system performs a critical coordination function between maintenance and operations. It should not offer too many choices or be too complex, but it should also not offer too few choices. Not only is the quantity of choices at issue, but the description of the choices in terms of time limits is and details is, too. Proper consideration of these issues will aid in proper scheduling.

Organizations typically are better at specializing than at coordinating. The expression “things fall through the cracks” is an example of coordination problems among specialized groups. The priority system is a critical coordination device between operations and maintenance. What new work must start immediately? What other new work is too urgent to wait until next week? If it can wait beyond this week, how long can it wait? The priority system addresses these questions. The maintenance group must try to follow the current schedule of work but also know when it should break the schedule. In addition, the advance schedule for next week must consider all the work in the backlog that is ready to go, but it cannot possibly do it all in a single week. A credible priority system helps sort all of this work.

A credible priority system should have more than three levels. While many proactive maintenance tasks originate from the maintenance group itself, operations also initiates many work requests. How should operations describe the relative urgency of this new work? Considering that a plant usually distinguishes between emergency work that must be started now and other urgent work that can wait but not until next week, that leaves work that can wait until next week.

This would mean a plant could simply have three broad levels of priority: emergency, urgent, and routine. However, a goal of having fewer than 20% of work requests be classified as emergency or urgent would mean that 80% of the work would have only a single classification level.

A single level consisting of 80% of the backlog is insufficient to guide maintenance, so add a priority level for work that can wait beyond next week. To have a truly credible priority system, a minimum of four levels is required; five is better. This system might classify work as emergency (now), urgent (complete this week), within two weeks, within one month, or longer than one month. A credible priority system needs enough levels to spread out the work adequately.

On the other hand, having too many levels or too much complexity also will cause problems. The more levels you add beyond five, the more difficult it will be for operations to make quick judgments, and such quick judgment is important in the context of coordination. Having more than five levels may make more people likely to classify their request as a level four or five, but at some point having a surfeit of levels becomes simply “gaming” the system, and the system will lose credibility. Extremely complex systems also hinder coordination in that operations will have difficulty knowing when the work will be completed, if at all.

The descriptions of the levels is another matter. Should the levels set time deadlines or should they describe the types of situations appropriate for each level? The problem with setting deadlines is that if there is too much work, maintenance will not be able to meet the deadlines. It is also difficult for operations to set deadlines for maintenance work. The advantage of qualitative descriptions is that whatever the volume of work, requests can be sorted in a logical order of relative importance. But it can be difficult to determine precisely how critical are the risks to efficiency, availability, personnel, and the environment associated with a given request. A system that has deadlines but also a limited number of qualifying descriptions might be helpful.

In summary: Try to avoid having too many or too few levels, and try to avoid having overly complex systems or items in the system that do not dictate the relative urgency of response. Remember that the point of the priority system is to help operations and maintenance quickly and easily communicate the relative urgency of the work.

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments