A maintenance shutdown planned ahead, with no product scheduled, is excluded time: it does not lower OEE, but it does lower TEEP. Maintenance done inside scheduled production time, and every breakdown repair, counts against availability.
The two clocks behind the answer
OEE is measured against scheduled time. TEEP is measured against all the time on the calendar, so nothing can be taken out of it.
So the real question is whether the maintenance sits inside the schedule. Here is one week on an extruder, shown as a waterfall.
The shutdown is booked well in advance, and no product is planned during it. In Overall Equipment Effectiveness (R. C. Hansen, 2001), a shutdown planned at least a week ahead is excluded time.
What that does to the numbers
The 5,400 good minutes stay the same whichever way you book the shutdown. Only the OEE denominator changes.
| Measure | Booked as excluded | Booked inside schedule |
|---|---|---|
| Scheduled time | 6,720 min | 7,200 min |
| OEE | 5,400 ÷ 6,720 = 80.4% | 5,400 ÷ 7,200 = 75.0% |
| Utilization | 6,720 ÷ 10,080 = 66.7% | 7,200 ÷ 10,080 = 71.4% |
| TEEP | 5,400 ÷ 10,080 = 53.6% | 5,400 ÷ 10,080 = 53.6% |
This is why TEEP is the honest check on capacity. TEEP = OEE × utilization, so any time taken out of OEE reappears as lower utilization, as shown in OEE vs TEEP and the TEEP calculator.
Sort any maintenance event in three questions
Excluded time. Not in OEE, but it lowers TEEP.
A planned stop in scheduled time (ST Operational). It lowers availability and OEE.
Downtime, usually DT Technical. It lowers availability and OEE, whatever the reason for the repair.
The full set of boxes is in the OEE loss tree. Common real cases then sort like this:
- Yearly overhaul, booked months ahead. Excluded time.
- Monthly service, booked on the weekly plan with no order on the line. Excluded time.
- Lubrication round in the middle of a shift that has product scheduled. A planned stop, so availability takes it.
- Operator checks before start-up. Excluded if the schedule starts after them, a stop if the schedule starts before them.
- A service moved in with two days' notice. It missed the one-week rule, so it counts as a stop.
- Planned shutdown that overruns by 3 hours. The plan is excluded, and the 3 hours are downtime.
- Repair after a failure. Downtime, always.
- An inspection with the machine running. It has no effect on OEE, since nothing stopped.
The cases that start the arguments
The three questions settle most events. These five are harder, so decide them once and write them down.
- Waiting for the technician. The wait is part of the same downtime, because the line is down from the failure until it runs again. Stamp the call and the arrival, so response time and repair time show apart, as in how to calculate MTTR.
- Slow running and scrap after a service. Excluded time ends when product is scheduled again. Whatever the restart costs then lands in performance and quality, and startup losses shows how to track it.
- Test runs after a repair. An unplanned test to prove the repair is part of the downtime. A trial planned ahead, whose product will not be sold, is excluded time.
- Service done during another stop. If a technician uses a changeover to grease a bearing, the minutes stay with the changeover. Each minute goes in one box, and the reason the line stopped decides which.
- One machine on a longer line. A service on one machine costs the line only if the line stops. With a buffer, the machine loses time and the line may lose none, as line OEE vs machine OEE explains.
Does more planned maintenance raise OEE?
It can, when the service prevents more stop time than it uses. Suppose the extruder loses 300 minutes a week to breakdowns, inside the 6,720 scheduled.
A 120 minute weekly service, booked ahead, removes 200 of those minutes. Outside the breakdowns, the line makes 5,400 ÷ 6,420 = 0.841 good minutes per scheduled minute, and we hold that pace.
Here both numbers rise, so the service pays for itself in output. If it prevented only 50 minutes, TEEP would fall, and OEE alone would not warn you because the 120 minutes are excluded.
That is the reason to read the two together before adding a service. Your stop history tells you which failures a service could prevent, and a downtime Pareto chart ranks them.
Why plants disagree about this
Some plants count all maintenance as a loss and want every minute visible. Others exclude all of it, and argue that a machine should not be punished for being looked after.
Both camps have a real worry. If maintenance lowers OEE, a crew under pressure may skip it, and if everything planned is excluded, the number gets inflated.
Our view is to use the rule above and show the excluded minutes next to OEE every week. That protects a planned service and keeps the time in plain sight, and how to set OEE targets covers keeping the goal honest.
The trap: moving routine work out of the schedule
Back to the extruder. Suppose 40 minutes of routine tasks each day, 200 minutes a week, are moved from stops to excluded time.
Scheduled time falls to 6,520 minutes and OEE rises from 80.4% to 82.8%, since 5,400 ÷ 6,520 = 82.8%. The extruder made nothing extra, and TEEP is still 53.6%.
Excluded time should be time you truly planned not to produce. Anything else belongs in the stops, as OEE calculation mistakes explains.
Worked case: the shutdown that overruns
Say the 8-hour shutdown takes 11 hours, so the line loses 180 minutes it was scheduled to run. At this line's usual pace of 0.804 good minutes per scheduled minute, that costs about 145 good minutes, leaving 5,255.
Booked correctly, the 480 stay excluded, the 180 are downtime, and OEE falls to 5,255 ÷ 6,720 = 78.2%. Book all 660 minutes as excluded instead, and OEE reads 5,255 ÷ 6,540 = 80.4%, as if nothing went wrong.
Where each number is useful
Use OEE to run the shift. It answers how well the line used the time you asked it to run.
Use TEEP when capacity is the question, for example before buying a machine or adding a shift. Utilization vs OEE explains how the two fit together, and the production capacity calculator turns it into units.
For long planned stops, report the hours as well. A maintenance plan that takes 480 minutes a week is 8 hours of capacity, and that is worth reviewing with TPM methods.
How to record it on the floor
- Mark the plan date. For each service, record when it was booked. That proves the one-week rule.
- Record planned and actual length. The plan goes to excluded time, and the overrun goes to downtime.
- Give planned stops their own code. A code for routine stops inside the schedule keeps them apart from breakdowns. See downtime reason codes.
- Review the split monthly. Rising excluded time with flat output means the rule is slipping.
- Enter it the same way each time. Planned time in your sum is the scheduled minutes after exclusions, and maintenance inside the schedule goes in as a stop. The OEE calculator takes those two inputs.
The difference between the two kinds of stop is in planned vs unplanned downtime.
Four ways this goes wrong
- Calling a breakdown repair planned because the part was on the shelf.
- Booking an overrun as part of the planned shutdown.
- Comparing OEE between plants without asking how they treat maintenance.
- Skipping a service to lift this week's number, when the service was never in OEE to begin with.
The split between planned and unplanned time only holds if each stop is recorded when it happens. Fabrico takes stops live from the machine and operators tag the cause at the line, so the minutes in OEE per line and shift come from the machine, including any overrun. See how Fabrico tracks stops.
What people ask next
Is preventive maintenance included in OEE?
Only when it happens inside scheduled time. A preventive service booked ahead, with no product scheduled, is excluded. A service in the middle of a scheduled run lowers availability.
Does planned maintenance lower TEEP?
Yes, always. TEEP counts all calendar time, so any hour the machine is not making good product lowers it. Check your own figures with the TEEP calculator.
Is cleaning treated the same as maintenance?
The same test applies: planned as not for production, or inside the schedule? Routine cleaning between runs usually sits inside the schedule. See sanitation downtime.
What about a planned shutdown that finishes early?
The machine may go back to work sooner, and if production restarts, the time from then on is scheduled. If nothing is planned on the line, the time stays excluded, since good work should never hurt OEE.
Does waiting for a maintenance technician count against OEE?
Yes, when it happens inside scheduled time. The wait is part of the same downtime, so availability takes all of it. Record the call and arrival times so you can see it, and use how to calculate MTTR for the split.
How much time should we allow for planned maintenance?
That depends on the machine and its plan, so we give no general figure here. The preventive maintenance calculator helps you work through your own numbers.
Fabrico is a light MES for OEE. It takes stops, speed and counts from PLCs or IoT sensors, with add-on AI cameras counting output where a machine has no usable signal, and operators tag the cause at the line.
Related guides: OEE vs OOE: The Difference and When to Use Each · OEE on a New Line: When to Start the Clock · OEE vs Technical Availability: The Key Gap · MTTR Reduction Plan: A 12-Week Template to Use