A slow season should barely move OEE, because OEE only counts the hours you schedule, while demand shows up in loading (scheduled hours ÷ calendar hours) and in TEEP (OEE × loading). If OEE falls when orders fall, first check whether idle hours are still sitting in the schedule.
Three numbers, three questions
Seasonal plants get confused because they ask one number to answer everything. Keep these three apart, and most of the confusion goes away.
| Number | The question it answers | Should it swing with demand? |
|---|---|---|
| OEE | How well did we run the hours we scheduled? | Not much |
| Loading | How many of the calendar hours did we schedule? | Yes, that is its job |
| TEEP | How much of all the hours became good output? | Yes, with loading |
Some reports call loading "utilization", and that word has other meanings too. Utilization vs OEE sorts them out, and OEE vs TEEP covers the definitions.
A year on one line
Example: a line runs 120 hours a week at peak, 80 in a normal period and 40 in the low season. The figures below are hypothetical.
| Season | Hours per week | Loading | OEE | TEEP |
|---|---|---|---|---|
| Peak | 120 | 71.4% | 72% | 51.4% |
| Normal | 80 | 47.6% | 75% | 35.7% |
| Low | 40 | 23.8% | 68% | 16.2% |
TEEP falls by two thirds from peak to low, while OEE moves only 4 points. OEE's widest gap is 7 points, from normal to low, and those points are worth a closer look.
Why OEE still moves a little
A small seasonal wobble is normal, and it usually has a real cause on the floor. Both directions are worth checking.
- Low season: short weeks. The same start-up costs less in a long week than in a short one. Example: five starts at 29 minutes are 145 minutes, which is 6.0% of a 40-hour week but only 2.0% of a 120-hour week. See startup losses.
- Low season: smaller orders. More changeovers per scheduled hour lower availability.
- Peak: new people. Temporary staff and overtime can raise stops and scrap, so check quality and stop rates by crew.
- Peak: deferred care. Skipped cleaning and maintenance can show up as stops weeks later.
- Any time: product mix. Seasonal products have their own rates, so use a rate per product or the mix will move the score.
The big trap: idle hours left in the schedule
If OEE collapses in the slow season, the line has rarely got worse. More often the crew's hours stayed in the schedule after the orders went away.
Time with nothing to run belongs outside OEE, when the plan says so in advance. The exact rules, including what to do when an order disappears on the day, are in how to treat time with no orders.
Update the schedule when the plan changes, not at year end. A shift calendar that still shows peak hours in February will quietly wreck your numbers.
Reading your monthly report
Put OEE and loading on the same page each month. The pair tells you which kind of change you are looking at.
| OEE | Loading | What it likely means |
|---|---|---|
| Steady | Down | Demand fell and the line runs as before |
| Down | Steady | A real line problem, or idle hours still in the schedule |
| Down | Down | Shorter weeks and runs, so more starts and changeovers per hour |
| Up | Down | Weak hours or hard products dropped out first |
The second row is the one to check first in a slow month, because hours left in the schedule hold loading up while they drag OEE down. Compare scheduled hours with the hours that actually had orders behind them.
The setup is small: one shift calendar per line, updated when the plan changes, and one code for time with nothing to run. Everything above follows from those two records.
How to compare across seasons
Comparing January with July tells you about the season. Three comparisons tell you about the line.
- Same season, last year. This cancels most of the mix and run-length effects.
- A rolling 12 months. Weight each period by its scheduled hours.
- Loss minutes per scheduled hour. Stops per 100 hours can be compared between a 40-hour week and a 120-hour week.
The weighting matters more in a seasonal plant, since the periods differ so much in size.
See how to average OEE for the reasoning.
Targets need the same care. A single yearly target will look easy in peak weeks and out of reach in short ones, so set one per season, based on the run lengths and start-ups each season really has.
Then judge each season against its own target and against the same season last year.
Use the numbers to plan the peak
This is where OEE earns its keep in a seasonal business. It tells you how many hours the peak will take.
Example: peak demand is 60,000 units a week and the ideal cycle time is 6 seconds, so the line needs 100 hours of perfect running.
At 72% the plan needs a weekend shift or stock built ahead. Raising OEE to 83.3% would fit it into the 120 hours you already run, and the production capacity calculator lets you try your own numbers.
The low season pays for that. Plan changeover practice, training and larger maintenance jobs into those free weeks, a week or more ahead, and they stay outside OEE as excluded time.
If demand is above capacity every year, read how to increase production capacity. The TEEP calculator shows how many hours you have left before you buy anything.
To tell a seasonal swing from a real loss, you need the same measure in every season. Fabrico shows OEE per line and shift, from stops, speed and counts taken live from PLCs or IoT sensors, with add-on AI cameras counting output where a machine has no usable signal. See how Fabrico shows OEE per line and shift.
What people ask next
Should OEE go down when demand goes down?
Not by much, if the idle hours are taken out of the schedule. A big fall usually means those hours stayed in, or that runs got much shorter. See OEE with no orders.
What is loading in OEE?
Loading is scheduled hours as a share of calendar hours. TEEP = OEE × loading, as in the TEEP calculator.
Should I use OEE or TEEP in a seasonal plant?
Use OEE to run the line and TEEP to plan capacity. A low TEEP in the slow season is expected, and it is a sales and planning topic. OEE vs TEEP compares them.
How do I set an OEE target if demand swings?
Set it by season, based on your run lengths and start-ups, and compare each one with the same season last year. How to set OEE targets shows the method.
Should we slow the line down when orders are low?
Run fewer hours at the accredited rate, and keep the ideal rate fixed. If you lower the ideal to match a slower pace, the trend stops meaning anything. Speed losses covers why.
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.