It is apparently common in business education to teach this concept. However, it is more complex than I have seen put into practice by controllers. In my practical experience,
it does not hold at all except at the extremes.
The answer will be very specific to your situation. You'd have to know about the kinds of machines, changeovers and shift schedules, and how much cleaning is counted as maintenance vice an operational task in your accounting system.
Coming from a continuous production line this was my experience on the department level:
At maximum production, there was minimum spending on maintenance. The production unit was fully scheduled, running efficiently without upsets. There was no time or need for serious maintenance (PM or CM). This was the baseline you refer to: only planned PMs and minor troubleshooting from people who were sunk cost anyway.
When production went down, many
bad things went UP: overtime, waste, scrap, intensive energy consumption, and maintenance costs (whether internal, external, or parts.) Basically, maintenance costs were incurred during maintenance periods, whether scheduled or not. During these periods, production was low, so the maintenance cost per unit produced got really high. I made this graph for my department and the correlation held over 2 decades. It was such an obvious lesson that I never updated that graph again. Instead, I felt silly for committing the time to collect all the data required for such an obvious lesson.
Another experience: We changed a unit from continuous 24/7 production to block 24/4 production, where they would operate 24 hours a day for 4 days straight. Operations labor when down, but repair and scrap went up. In this case, cooling extruders down for 3 days, then warming them up too aggressively caused all kinds of problems that weren't evident in continuous production. Example: trying to get started quickly on Monday morning, don't heat up for long enough, blow a rupture disc because the polymer is too viscous. We went from almost no rupture disc consumption to nearly one per extruder per week...so we ran out. This was not the only problem of this sort. It took about 2 years to learn how to operate with frequent startups and change SOPs. So in this case, lower production meant higher costs again.
The only time when maintenance costs went down as production went down was when a production unit was intentionally mothballed for a period of months. Even in this case, special effort was needed to restart the equipment: recalibrate instruments, re-baseline vibration levels, discover new leaks. I did a little work trying to correlate the re-investment to the length of downtime. However, there were many variables, and I didn't have many scenarios to work with. The quality of the mothball preparation was a huge factor. I found a mixer running in a dry vessel. For water systems, keeping pumps on and chemistry control in spec...having a person rotate big bearings weekly...these became part of the baseline cost for the rest of the operating plant. This looked on paper like our maintenance cost per unit went up, so it was not smart but it was wise. Future production orders got prevention for free.
The maintenance cost/production volume correlation might be different on a macro level. When I was doing ship maintenance management for the navy at a fleet headquarters, we had a similar discussion. We could not demonstrate from our records that incrementally more maintenance spending would result in higher operational availability for one individual ship. On a bigger scale, we could show that maintenance spending under the lifecycle baseline would result in lower operational availability, but it was over years and dozens of ships. The pattern was only apparent at scale.
------------------------------
Karl Burnett
General Electric
Anderson SC
------------------------------
Original Message:
Sent: 12-15-2022 01:27 PM
From: Timur Celebi
Subject: Adapting Total Maintenance Budget to Changing Volume
Hello,
I am looking for some good practices for adapting total maintenance budget with regards to changing volume (total produced goods) in FMCG industry (metal packaging).
Obviously there are some fixed maintenance activities that need to be carried out regardless of the volume and some variable linked with volume and it is not easy to find out what percentage of total maintenance cost/activity is variable what percentage is nonvariable.
Moving from time based to run time based, relying more on advanced methodologies like RCM, having PdM guided proactive maintenance can help to adapt to changing volume.
But is there a rule of thumb that you are using for adapting maintenance budget to changing volume?
Regards,
Timur Celebi