Implementing a priority system to create and maintain a profitable plant
Recently we discussed using a simple, but not too simple, priority system to coordinate between operations and maintenance. This month let’s go past the workings of the system and more into the philosophy. The priority system is a key player in facilitating our profitability. But who should care? Who sets the work priority? Should it ever change? And how well are we using the priorities?
Planning a fully loaded weekly schedule at your plant
Why a weekly schedule is the right timeframe for productivity
Implementing a priority system to create and maintain a profitable plant
Quick recap: The Palmer preferred priority levels are something along the lines of 0-Emergency, 1-Urgent, 2-Next Week, 3-This Month, and 4-Longer. Five levels allow two levels for breaking the weekly schedule (perhaps 20% of our work) and three levels for can wait a week, can wait a bit longer, and don’t want to forget (altogether about 80% of our work).
Priority system support and understanding
The priority system is a huge deal! Our spread of work per the priorities actually defines the profitability of our plant! We make profits with assets that run to produce a product we sell. The more priority 0’s and 1’s (emergency and urgent) work orders we generate, the less our assets are running. Not every breakdown threatens production. For example, a primary pump fails, the backup pump starts, and the operator writes a priority 2 or 3 (non-urgent) work order. But no plant has perfect designs that cover all contingencies and people write priority 0 and 1 work orders when situations threaten production. Having higher profitability is absolutely related to having fewer priority 0 and 1 work orders.
Management and first line maintenance supervisors must care. Priority 0 and 1 work orders are defects! Maintenance is about preventing emergencies and urgent situations that threaten production. Having 0’s and 1’s mean we have “failed to maintain.” Now we are in the “fix it” mode recovering production assurance. Management must not be comfortable simply that we fix 0’s and 1’s as soon as possible. If we average, say, 600 new 0’s and 1’s each month, management must care about reducing that rate of defects. First line maintenance supervisors must also care. These supervisors must push back against operators that call wanting this or that done. It’s no problem that operators call, but supervisors must insist operators use the appropriate priority code, 0 for break the schedule today, and 1 for break the schedule this week. We must be able to quantify our defects.
Generally, the original requester of the work sets the priority of the request. This operator is on the spot seeing the exact asset condition and understands the importance of the asset in its operating context: “Hey, if we don’t fix this valve today, we’ll lose too much product” or “This motor can wait a week or so.”
We also need to help everyone understand the priority system. Consider bulletin boards and putting little business cards on everyone’s keyboard showing the levels with a few words of description. Have the CMMS computer default new requests to priority 3 or 4. Some plants have a gatekeeper that is usually a senior, experienced operator helping judge the new priorities before maintenance has to scramble: “That broken pump can really wait until next week” or “The overall water supply is low and the demineralizer must be fixed tomorrow.” Some plants have a brief morning meeting to review all new work requests with the same function as a gatekeeper.
Priority system accuracy
Consider also the accuracy of the priority selection. It is often difficult for a requester to decide “Is this really a 1 or 2?” “Is this really a 3 or 4?” People with unproductive maintenance forces are very concerned with picking the correct priority “because 4’s will never get done.” But people with productive maintenance forces are not overly concerned with the priorities because everyone’s work gets done.
Generally, we should not change the priority after a work order has arrived in maintenance. Don’t change a priority 0 or 1 to a priority 3 because we won’t have the right parts on-site for a few weeks. Instead, keep the priority and be able to measure how well the storeroom supports emergency and urgent work. We also should not “age” work orders across priorities: Some people say that a priority 4 should become a priority 3 after so long, then a priority 2 after so long, etc. But a priority 4 squeaky wheel will never be as important as a priority 1 boiler feed pump banging loudly right now. Instead, we need to be more productive to complete the priority 4 work. Nonetheless, a gatekeeper properly should raise the wheel from a 4 to a 1 if the squeak has gotten worse and may soon impact production.
Measuring planning and maintenance
Evaluate how we are doing, overall for maintenance and in the planning phase. Measure if we usually finish priority 1’s within a week of being requested, priority 2’s within a couple of weeks, priority 3’s within a month, and at least know how we do on priority 4’s. Use a “median” score rather than a “mean” average. (A straight mean average would cause one or two work orders that took forever to skew the meaning, no pun intended.) For planning, we should generally plan the work in half the priority allotted completion time. In other words, try to plan the priority 2 work (2 weeks to complete) within a week; plan the priority 3 work (a month to complete) within two weeks; and plan the priority 4 work within a month. We can also measure medians for planning times.
The priority system is a super big deal: Managers, supervisors, operators, and planners must actively involve themselves in leveraging the priority system to create and maintain a profitable plant. “That’s what it’s all about.”
This story originally appeared in the May 2022 issue of Plant Services. Subscribe to Plant Services here.
This article is part of our monthly Palmer's Planning Corner column. Read more from Doc Palmer.
