In reactive environments, production will often issue annual schedules for the total operation, dictating the allowable time for shut-downs. This was the case at a plant where I worked several years ago. The special coating application department work center was the key asset to the entire plant site; all product lines had to sequence their product through the special process. This production area was the lowest cost producer for a multinational company. The corporate strategy was to use this machine to produce as much volume as possible. Consequently, production usually worked 24 hours a day, 7 days a week.

The special operation was very large. It occupied a large three-story building. The machine's continuous product length was more than half a mile long. The special machine system was also targeted for major programs to speed up, to improve process quality, and to upgrade several areas of technology. Significant project work was to be part of each shut-down.  The process was classified as a continuous discrete system. Major changeovers were sometimes four days apart.

The existing maintenance plan when I began was to schedule a pair of two-week shutdowns each year. One shutdown was early in the year, the other one late because of summer seasonal volumes of some products. The work area's experience with startups after shutdowns was very poor. Therefore, the fewer shutdowns that were needed, the better. As you may suspect, overall equipment reliability and availability were not as good as desired. Maintenance and production had an "us versus them" relationship with each other. In between shutdowns the maintenance approach was to fix the equipment only when it broke. The only condition monitoring method used was tribology testing of oils from large gearboxes.