Master production schedule (MPS): a worked example and an interactive calculator

What a master production schedule is

The MPS sits between strategy and execution. Above it, sales and operations planning agrees monthly volumes for product families. Below it, MRP and the shop floor need exact quantities of exact products on exact dates. The master scheduler turns the first into the second and keeps it realistic as orders arrive.

Where the MPS sits in production planning

The MPS horizon must be at least as long as the cumulative lead time of the product (the longest path of purchasing and manufacturing lead times through the bill of materials). Anything shorter and MRP will generate orders that are already late.

The MPS record, line by line

• Forecast — expected demand for the period.
• Customer orders — demand already booked.
• Projected available balance (PAB) — stock expected at the end of the period.
• MPS quantity — the lot scheduled for completion in the period.
• Available to promise (ATP) — the part of stock and scheduled production not yet committed to customers.

demand(t) = customer orders(t)                   inside the demand time fence
          = max(forecast(t), customer orders(t))  beyond it

PAB(t)    = PAB(t−1) + MPS(t) − demand(t)
MPS(t)    = lot size × k, the smallest k that keeps PAB(t) ≥ safety stock

ATP(first period) = on hand + MPS(1) − orders until the next MPS quantity
ATP(t, MPS > 0)   = MPS(t) − orders from t until the next MPS quantity

Worked example

From the baseline schedule above, with 45 lamps on hand and lots of 60:

1. Week 1 is inside the 2-week demand time fence, so only the 38 booked lamps count, not the forecast of 40. PAB = 45 − 38 = 7. No lot is needed.
2. Week 2: demand is 32. 7 − 32 would go negative, so the MPS schedules one lot of 60, and PAB ends at 35.
3. Week 3 is the first week beyond the fence: demand is the greater of the forecast (40) and orders (20), so 40.
4. ATP in week 1 = 45 on hand + 0 − 38 booked before the first lot arrives = 7.
5. ATP in week 2 = the 60-lamp lot minus orders booked from week 2 until the next lot in week 3 = 28.

Try the rush order preset: an extra 50 lamps in week 2 force a new lot inside the demand time fence. The arithmetic accepts it. Whether the plant can, given materials already ordered and capacity already allocated, is the master scheduler's judgement.

Available-to-promise and order promising

ATP is what lets sales answer “can you deliver 50 by week 3?” without guessing. Two conventions are common:

• Discrete ATP (used here) is calculated per MPS lot, covering orders until the next lot. A negative bucket borrows from the one before it.
• Cumulative ATP adds discrete ATP forward through time, so each period shows everything that could still be promised up to then. To promise an order, find the first week in which cumulative ATP covers it, which is what the calculator's promise check does.

ATP and projected available balance answer different questions. PAB subtracts expected demand (forecast or orders) and answers “will we run out?”. ATP subtracts only booked orders and answers “what can sales still commit?”. Where the forecast is larger than the orders booked so far, ATP is higher than PAB. That gap is forecast demand that has not been sold yet.

Time fences

• Demand time fence (DTF) — inside it, only customer orders count and the forecast is ignored. Changes here are expensive: materials are committed and capacity allocated.
• Planning time fence (PTF) — inside it, the system does not create or move MPS lots on its own. Only the master scheduler does. Beyond it, the system may plan freely.
• Frozen zone — some firms freeze the first weeks entirely; changes need management sign-off.

Fences exist to stop nervousness: without them, every new order would ripple straight through MRP into purchasing and the shop floor.

What the MPS schedules: make-to-stock, assemble-to-order, make-to-order

• Make-to-stock — the MPS schedules finished products, as in the lamp example.
• Assemble-to-order — when options multiply (a lamp in 6 colours with 3 bases is 18 end items), the MPS is set for modules using planning bills, and a separate final assembly schedule builds what customers actually order.
• Make-to-order — the MPS may schedule key materials or capacity rather than products that do not exist until an order arrives.

Checking the MPS: rough-cut capacity planning

A schedule that satisfies demand may still be impossible to build. Rough-cut capacity planning multiplies MPS quantities by the load each product places on critical resources (a bottleneck line, a test bench, skilled labour) and compares the result with available capacity, period by period. If week 4 needs 130% of the paint line, the MPS changes before MRP explodes it, not after.

Common mistakes in MPS exercises

1. Using the forecast inside the demand time fence instead of booked orders only.
2. Forgetting on-hand stock in the first ATP bucket, or adding it to later buckets.
3. Netting ATP against forecast. ATP only subtracts booked customer orders.
4. Scheduling a partial lot when the rule is a fixed lot size (or multiples of it).
5. Mixing up discrete and cumulative ATP when promising an order.

Teaching the MPS: exercises to set with this calculator

1. Promise or refuse. A customer wants 70 lamps in week 2. What can you promise, and when? Write the reply to the customer.
2. Lot size trade-off. Compare lots of 40, 60 and 120. How do the number of lots, average PAB and ATP change?
3. Move the fence. Set the demand time fence to 0 and then 4 weeks. Which weeks' lots change, and why does that matter to purchasing?
4. Rush order judgement. In the rush scenario, argue for and against accepting the order at the cost of a new lot inside the fence.

Frequently asked questions

References and further reading

• Jacobs, F. R., Berry, W. L., Whybark, D. C., & Vollmann, T. E. (2018). Manufacturing Planning and Control for Supply Chain Management: The CPIM Reference (2nd ed.). McGraw-Hill.
• Proud, J. F. (2007). Master Scheduling: A Practical Guide to Competitive Manufacturing (3rd ed.). Wiley.

The calculator implements the standard MPS record and discrete ATP with backward borrowing described in these texts.