How Far in Advance Should You Schedule Maintenance?
How far in advance should you schedule maintenance? New things go into the backlog constantly, so, hopefully as early as possible; however, a plant can only schedule as much work as it can do.
There are multiple schedules in maintenance, each with its proper purpose. Different horizons include a daily schedule, weekly schedule, monthly schedule, yearly schedule, and schedules beyond the current year. A plant should utilize all of these schedules: daily, weekly, monthly, yearly, and beyond.
There is also the “run-to-fail strategy,” without any scheduling, where a plant just fixes something after it breaks. Some of these schedules are better understood than others, but using each schedule correctly enhances maintenance effectiveness.
First, “run to failure,” also described as “don’t touch it if it isn’t broken; fix it when it breaks.” However, maintenance effectiveness is the primary objective. Maintenance is “keeping something from breaking.” A plant wants to maintain that something’s performance. If that something breaks, the plant has not done its job. The plant did not maintain it. No matter how fast the plant fixes it after it breaks, it’s better to keep it from breaking in the first place.
Real financial profit comes from keeping assets running. Downtime is expensive. Industry wisdom says the work done to keep assets from breaking is generally worth 10 times its value. In other words, every $100 worth of labour and parts the plant puts into an extra proactive maintenance task saves the plant $1,000 on the bottom line.
(Nevertheless, “run to failure” is a proper component of an RCM (reliability centered maintenance) strategy, but only going through a disciplined series of questions concerning how to keep something from breaking.)
The purpose of yearly maintenance schedules and multi-year maintenance schedules generally revolves around budgets, projects, and contractors. Projects include installing new assets or systems and modifying them, but also larger maintenance events such as specific planned major shutdowns and turnarounds. Schedules for a year generally show the exact dates of events, whereas schedules spanning multiple years are a bit more concerned with figuring out the dates of the events. They also involve writing specifications, gathering funding and bidding work.
To co-ordinate their preparation work, engineers and project managers need to know when the major events are scheduled. Some plants could do better on a monthly basis, updating their yearly and beyond-outage schedules. The yearly schedules would have more exact dates and the 10-year schedule might just list the events, but not exact dates. Since issues with yearly and beyond schedules are important, most plants do them relatively well.
A closely associated schedule with the yearly schedule is a particular shutdown schedule. This schedule for a particular event would include everything from the specifications to the bidding to the execution to the start-up. For larger outages, preparation work may start a year and a half before the start of the execution. Taking a production unit offline impacts profits, so plants usually perform these events.
Plants also manage, in concept, what might be called a monthly maintenance schedule. The purpose of looking at maintenance monthly is keeping up with preventive maintenance (PM). PMs are calendar- or usage-based maintenance tasks such as lubrication or human sensory inspections that plants perform on a regular basis. Plants want high PM compliance. The plant’s reliability function designs these PMs to keep things from breaking. The CMMS should automatically issue PMs into the work order backlog in advance.
An associated practice is predictive maintenance (PdM), which uses technology such as vibration analysis, ultrasound, or infrared thermography to predict an impending failure based on an asset’s current condition. PdM technicians complete routes with their technology to find otherwise undetectable indications of developing problems. As with PM, plants want high compliance with completing PdM routes to find small problems they can address proactively before the plants suffer breakdowns. Most plants correctly use monthly schedules to look at PM compliance, but many could do better keeping up with PdM route compliance.
The purpose of the daily schedule is to assign specific work orders to specific individuals and co-ordinate lock out/tag out with operations for specific times. This schedule presumes the existence of maintenance first-line supervisors. It should not be the responsibility of craftspersons to decide which work orders they should do next. There is a lot of churn in daily maintenance, with jobs taking longer or shorter than expected and with a significant amount of new reactive work arising that cannot wait. The supervisor looks at how current jobs are progressing and what work is waiting and then matches the appropriate individuals to the appropriate jobs.
The supervisor also co-ordinates this information with operations personnel for clearance to work on equipment. This understanding of current progress and deployment to the next jobs merits a supervisor, often full-time for a crew of 10 to 15 craftspersons. Usually in the early afternoon each day, the supervisor creates the daily schedule for the next day, showing expected assets needing clearance and assignments for individuals. In the later afternoon, the supervisor discusses the schedule with operations to request clearance.
Finally, the weekly maintenance schedule is the most misunderstood device for helping maintenance. Because of this confusion, it might provide the greatest opportunity for achieving more effective maintenance. On the one hand, the weekly schedule is not simply sticking five daily schedules together in advance. There is too much churn in real life maintenance. Such a schedule would require a great amount of time every single day, moving everything around for the rest of the week. On the other hand, the weekly schedule must exist.
Often plants will not have a weekly schedule at all. They reason that supervisors know about the work orders in the backlog, including PM and should keep everyone busy. The culture at plants with no weekly schedule is “we need to take care of operations and otherwise keep everyone busy.” The problem with this approach is the management question How much work should the workforce complete? The answer is not an indefinite “as much as it can with the supervisors keeping everyone busy.”
Industry experience has shown that plants that create a weekly schedule complete more work orders each week. The plant has a scheduler sort through the backlog, finding the best work for the crew the next week. The key is that rather than a complex assigning of individuals a week ahead of time, the weekly schedule is simply a batch of work that matches the whole crew capacity. The batch of work functions as goal setting to give a 10-person crew 400 hours’ worth of work orders for the next week.
The culture becomes “we need to complete this work, but it’s okay if we don’t do it all if operations suddenly needs us.” This definite subset of the backlog gives the supervisor a better focus on how much work to assign and surprisingly leads to as much as 50 per cent or more work completion–considering that most plants are already very good at handling all the reactive work from operations calls, that nearly all of the additional work completed through proper weekly scheduling is proactive work. Weekly scheduling of full labour capacity simply as a batch of work gives the plant a more effective maintenance program because the workforce starts to do more proactive work that soon heads off more breakdowns.
Rather than to address “how far ahead,” a single week seems to be better for the goal-setting aspect of productivity than going two or three weeks in advance. A plant might try to create advance schedules beyond a week as skeleton schedules, but not yet fully load them. Things do change at plants from week to week, especially with PdM finding things that can wait, but not for long. Operations calls can also sometimes be put off, but not for long. Fully loading the labour capacity for a single week seems to balance putting enough work together for a goal and allowing for some break in reactive work.
Each of these schedules has its place and is equally important, but the daily and weekly schedules tend to be the least understood. The daily schedule needs an active first-line supervisor. The weekly schedule needs to focus that supervisor on how much work.
Doc Palmer, PE, MBA, CMRP, is the author of McGraw-Hill’s Maintenance Planning and Scheduling Handbook. As Managing Partner of Richard Palmer and Associates, he helps companies worldwide with planning and scheduling success. For more information, including a schedule of current public workshops, visit www.palmerplanning.com, or e-mail Doc at firstname.lastname@example.org.