Monthly output is rarely the press speed multiplied by the calendar hours. Machines stop, speeds are limited by tablet quality, cleaning takes time, and the slowest stage in the process decides the final output. The calculation below runs from the machine rating to a realistic monthly figure.
All figures are illustrative. Substitute the press, OEE data and shift pattern of your own plant.
Watch: Compression machine capacity calculation
Step 1: the theoretical output of the press
On a single-sided press each die makes one tablet per turn of the turret. On a double-sided press each die makes two, one at each compression station. Check the data sheet of your own machine for the number of dies and the tablets made per turn.
Step 2: use the validated speed, not the maximum
The speed written on the machine plate is a mechanical limit. The speed that counts is the highest speed at which the product was validated, where weight variation, hardness and appearance stay inside their limits. A blend with poor flow may only run well at a much lower speed. The Bulk and Tapped Density Calculator and the weight variation guide help you see whether flow is the limit.
Step 3: apply OEE
Overall equipment effectiveness describes how much of the theoretical output turns into good product. It multiplies three factors:
- Availability. The share of planned production time that the machine was actually running, after breakdowns and long stops.
- Performance. The actual speed compared with the ideal speed, after slow running and small stops.
- Quality. The share of tablets made that are good, after rejects.
Good tablets per hour = Theoretical output × OEE
Measure the three factors from your own records. A figure copied from another plant, or from a presentation, will not describe your press.
Step 4: from tablets per hour to hours per batch
Divide the batch size by the good tablets per hour to get the running time. Then add the time that is outside OEE, such as changeover, cleaning and line clearance between batches. Be careful not to count the same time twice: if your OEE already treats changeover as lost availability, do not add it again.
Worked example: a 35-die press
A single-sided press has 35 dies and a validated speed of 30 rpm. Availability is 80%, performance 90% and quality 98%. The batch size is 400,000 tablets, the changeover and cleaning take 4 hours, and the plant plans 400 hours of compression a month (two 8-hour shifts over 25 working days).
| Step | Calculation | Result |
|---|---|---|
| Theoretical output | 35 dies × 1 × 30 rpm × 60 | 63,000 tablets/h |
| OEE | 0.80 × 0.90 × 0.98 | 70.56% |
| Effective good output | 63,000 × 0.7056 | 44,453 tablets/h |
| Running time per batch | 400,000 ÷ 44,453 | 9.0 h |
| Changeover and cleaning | Illustrative | 4.0 h |
| Cycle per batch | 9.0 + 4.0 | 13.0 h |
| Batches in 400 h | 400 ÷ 13.0 | 30.8, so 30 whole batches |
Compression alone could make 30 batches. Whole batches are the right unit, because a partial batch does not leave the plant.
Step 5: check the other stages
A tablet is not finished at the press. It still has to be coated, packed and released, and the slowest of those stages sets the limit for the month.
| Stage | Time per batch | Batches in 400 h |
|---|---|---|
| Compression | 13.0 h (9.0 run + 4.0 changeover) | 30 |
| Coating (one pan) | 14.0 h | 28 |
| Blister packing | 10.4 h (7.4 run + 3.0 line clearance) | 38 |
| Plant output | Set by the slowest stage | 28 |
In this example the single coating pan is the bottleneck. The press could make 30 batches, but the plant can only finish 28. The extra two batches would pile up in front of coating as work in progress. Options include a second shift on the coating pan, a shorter coating cycle, a second pan, or running the press fewer hours and using the saved time for maintenance. Tablet load and spray set-up affect the coating cycle, as explained in Coating Pan Loading.
Interactive calculator
Quick calculator (reference only)
Use the validated turret speed, not the machine maximum. If your OEE already counts changeover as lost availability, set changeover to 0.
From batches per month to the year
Twenty-eight batches a month is 336 batches over twelve months, before planned shutdowns, validation runs and holidays. That yearly figure feeds directly into the cost work in What a 1% Yield Drop Really Costs, because the same percentage of yield is worth more on a larger number of batches. Material lost at the press is tracked in Compression Reconciliation, and the pressure used is explained in Compression Force to Pressure.
Planning errors
- Using the maximum turret speed instead of the validated speed.
- Planning with an OEE that was never measured, or one taken from another plant.
- Counting changeover time twice, once in OEE and once as a separate addition.
- Rounding batches up. Only whole batches count.
- Planning the press in isolation and ignoring coating, packing and release.
- Forgetting that the OEE and the changeover time differ between products, especially for products with long cleaning.
- Using calendar hours instead of the hours the press is actually scheduled to run.
Questions on monthly output
How do you calculate tablet press output per hour?
Multiply the number of dies by the tablets made per die per turn and by the turret speed in rpm, then by 60. For a single-sided press, that is dies × rpm × 60.
What is a good OEE for a tablet press?
It depends on the product, the batch length and how long changeovers take. The useful comparison is your own press over time, not a general target.
Why is the plant output lower than the press output?
Because the slowest stage in the process sets the limit. Coating, packing or release often limits output before the press does.
Should the changeover time be included in the batch time?
Yes, unless your OEE figure already counts it as lost availability. Count it once, and say which method you used.