Turn every product into minutes: OEE = Σ(good units × that product's ideal cycle time) ÷ scheduled time. Do not use one average ideal rate for the whole line, because then OEE moves whenever the product mix moves, even if the line ran exactly the same.
Mixed product OEE calculator
Enter one shift on one line. The defaults are the two-product shift worked through below. Updates as you type.
Ideal minutes per product
Rate per product or one average rate
The working
Quality is weighted by ideal time, so a scrapped slow unit counts for more line time than a scrapped fast one.
The problem with counting units
A carton of 12 and a carton of 48 do not take the same time to make. Adding them as "two cartons" hides that.
Units from different products cannot be added in a useful way. Minutes of ideal running can, and that is the whole trick.
This is the same rule as the single-product formula. It just runs once per product, and the performance calculation works the same way.
Worked example: two products in one shift
Example: a line is scheduled for 480 minutes and loses 60 minutes to stops, including one changeover. That leaves 420 minutes of run time, split between two products.
| Product A | Product B | Total | |
|---|---|---|---|
| Ideal cycle time | 2 s | 4 s | |
| Run time | 240 min | 180 min | 420 min |
| Units made | 6,000 | 2,400 | 8,400 |
| Good units | 5,880 | 2,340 | 8,220 |
| Ideal minutes, all units | 200 | 160 | 360 |
| Ideal minutes, good units | 196 | 156 | 352 |
Now the three factors work as usual, with minutes in place of units.
| Factor | Sum | Result |
|---|---|---|
| Availability | 420 ÷ 480 | 87.5% |
| Performance | 360 ÷ 420 | 85.7% |
| Quality | 352 ÷ 360 | 97.8% |
| OEE | 352 ÷ 480 | 73.3% |
Check it: 0.875 × 0.857 × 0.978 = 73.3%. The short route, 352 ÷ 480, gives the same answer.
Watch the quality line. Counting units gives 8,220 ÷ 8,400 = 97.9%, which is slightly off, because a scrapped slow unit costs more line time than a scrapped fast one.
What goes wrong with one average rate
Teams with many products often use one rate for the whole line. Each choice fails in its own way, using the same shift as above.
| Single ideal for all 8,400 units | Ideal minutes | Performance |
|---|---|---|
| A's rate, 2 s | 280 | 66.7%: B looks like a slow line |
| Average, 3 s | 420 | 100%: every speed loss vanishes |
| B's rate, 4 s | 560 | 133%: above the maximum |
| Per product (correct) | 360 | 85.7% |
The real damage comes the next day. A single rate turns a change of plan into a change of score.
Example: on day 2 the same line runs A for 120 minutes and B for 300 minutes, at the same speed on each product as day 1. That makes 3,000 of A and 4,000 of B, or 7,000 units in all.
Day 2 by product: A is 3,000 × 2 s = 100 ideal minutes and B is 4,000 × 4 s = 266.7, so performance is 366.7 ÷ 420 = 87.3%. Day 2 by single rate: 7,000 × 2 s = 233.3 ideal minutes, and 233.3 ÷ 420 = 55.6%.
Meetings about a fall like that one waste everyone's time. The line did nothing different; the planner did.
Which ideal cycle time does each product get?
A rule here matters more than the sums: one accredited rate per product family of the same size and format.
This is the method in Overall Equipment Effectiveness (R. C. Hansen, 2001):
- A real difference in format earns its own rate. A 2-litre bottle and a 500 ml bottle are different families, with different proven speeds.
- A product that is simply harder to run does not. If a sticky recipe in the same format runs slower, keep the family rate and let OEE drop.
That drop is useful information, because it shows what the product costs in line time. The rules for setting each rate are in ideal cycle time.
Where changeovers go
A changeover between products is a stop, so it lowers availability for the line, as how to count changeovers in OEE explains. It belongs in the line total whatever else you do with it.
If you also split the line by product, someone has to carry it, and practice varies. We charge the incoming product and keep that rule fixed; OEE by product covers the split and the other options.
How to run it in a spreadsheet
Keep one row per run, with four columns: scheduled minutes, product, good units and ideal seconds per unit. Pull the ideal seconds from one fixed product table so nobody types them by hand.
The OEE tracking template gives you a layout to start from.
Mistakes to check for
- Mixed units. Packs per minute and cartons per minute look alike in a report. Write the unit next to each rate.
- A moving bottleneck. A product can be limited by the filler while another is limited by the packer. Use the rate of the step that limits that product.
- Product table drift. An edit to one rate changes every past comparison, so date each change and mark it on the chart.
- Averaging product OEEs. Weight them by scheduled time, as in how to average OEE across machines.
- A new product with no rate. Do not leave it out quietly. Give it a provisional rate, mark it as provisional and replace it once proven.
Once the sum is right, the six big losses are comparable from week to week. A change in the score then means a change on the floor.
A mixed line needs counts and speed tied to what was running at the time. Fabrico takes stops, speed and counts live from PLCs or IoT sensors, and shows OEE per line and shift. See how Fabrico shows OEE per line and shift.
What people ask next
Can I use a weighted average ideal cycle time?
Only if you weight it by the units made in that same period, which is the sum above in another form. A fixed average from last year stops being true as soon as the mix changes. Ideal cycle time explains what to use.
Why did OEE drop when we ran more of a slow product?
With a rate per product, a slow product does not lower OEE by itself. If OEE still fell, look for more changeovers or a real speed loss, or check that the rate is per product. OEE calculation mistakes lists more checks.
Should a hard product get a lower ideal rate?
Not if it has the same size and format as its family. Keep the family rate so the time the product costs stays visible. Use OEE by product to see which ones cost most.
How do I calculate quality with several products?
Weight it by ideal time: good ideal minutes ÷ all ideal minutes. That way the three factors multiply back to the same OEE. How to calculate OEE quality covers scrap and rework.
How should I handle hundreds of SKUs?
Group them into families by size and format, each with one rate. A line with 300 SKUs may need far fewer rates than that, and the sum stays the same.
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: Ideal Run Rate in OEE: How to Calculate It