Master production schedule

Why it matters

The master production schedule (MPS) is the plant's promise: how many of each end item will be completed in each week. It turns the aggregate plan into specific products, drives MRP for every component and purchased part, and tells sales what can still be promised to customers. A wrong or unstable MPS ripples down into shortages, expediting and idle capacity.

Key ideas

What the MPS is. A time-phased statement of the quantity of each end item (or, for assemble-to-order products, each major module) to be completed per period, usually per week, over a horizon at least as long as the cumulative lead time of the product. It disaggregates the aggregate plan: the sum of MPS quantities for a family should match the aggregate plan for that family.

What is master-scheduled.

• Make-to-stock: finished products (few end items made from many components).
• Assemble-to-order: modules or options, because the number of end-item combinations is huge; a final assembly schedule (FAS) assembles them when orders arrive.
• Make-to-order: often the purchased or long-lead raw materials and the key end items.

Inputs. Opening inventory, forecast by period, booked customer orders, lot-sizing rule, safety stock, and capacity limits.

The MPS record (per item).

• Requirements: in each period take the larger of forecast and booked customer orders (near-term periods are mostly orders; later ones mostly forecast).
• Projected on-hand (POH) inventory: opening POH + MPS quantity − requirements.
• MPS quantity: whenever POH would fall below zero (or below safety stock), schedule a lot (fixed lot size, or lot-for-lot) in that period.
• Available-to-promise (ATP): the part of on-hand stock and scheduled MPS lots not yet committed to customer orders. Sales uses it to quote delivery dates.

ATP (discrete method).

• First period: ATP = on-hand + MPS in period 1 − customer orders booked until the next MPS receipt.
• In any later period that has an MPS quantity: ATP = MPS quantity − customer orders booked from that period up to (not including) the next MPS receipt.
• Periods without an MPS quantity have no ATP entry. If an ATP comes out negative, it is covered from the ATP of an earlier period.
• Note that ATP uses only booked orders, never the forecast, because it answers "what is still free to promise".

Time fences. Changing the MPS close to the due date disrupts purchases already placed and work already started.

• Frozen zone (inside the demand time fence, about the final assembly lead time): no changes without senior approval.
• Slushy (firm/trading) zone (up to the planning time fence, about the cumulative lead time): changes allowed if materials and capacity can be traded.
• Liquid (free) zone: beyond the planning time fence, any change is allowed; the system can replan automatically.

Rough-cut capacity planning (RCCP). Before the MPS is released, the load it puts on critical (bottleneck) resources is checked using bills of resources (hours per unit at key work centres). If load exceeds capacity, the MPS is levelled or capacity is added. The detailed check, after MRP, is capacity requirements planning (CRP).

Stability vs responsiveness. Frequent changes ("MPS nervousness") cause expediting and lost credibility; too rigid an MPS ignores real demand. Time fences and reasonable lot sizes balance the two.

Formulas

Requirement(t) = max[F(t), CO(t)]

• F = forecast (units), CO = booked customer orders (units), t = week.

POH(t) = POH(t−1) + MPS(t) − Requirement(t)

• POH = projected on-hand inventory at the end of week t (units), MPS = master scheduled receipt in week t (units). Start with POH(0) = current on-hand.

Schedule MPS(t) = lot size Q whenever POH(t−1) − Requirement(t) < SS

• Q = lot size (units), SS = safety stock (units, zero if none). If one lot is not enough, schedule as many lots as needed.

ATP(1) = OH + MPS(1) − Σ CO(1 … k−1) ATP(k) = MPS(k) − Σ CO(k … m−1)

• OH = current on-hand (units), k = a week with an MPS receipt, m = the next week with an MPS receipt (or end of horizon + 1).

Load on resource j = Σ (MPS quantity of item i × hours per unit of i at j)

• Rough-cut check against available hours at j.

Worked examples

Example 1 (standard) – MPS and ATP without safety stock. On-hand = 60 units, lot size = 150 units, no safety stock. Forecast, weeks 1–8: 40, 40, 40, 40, 60, 60, 60, 60. Customer orders, weeks 1–8: 50, 35, 20, 10, 5, 0, 0, 0.

1. Requirements = max(F, CO): 50, 40, 40, 40, 60, 60, 60, 60.
2. Week 1: POH = 60 − 50 = 10 (no MPS).
3. Week 2: 10 − 40 < 0, so MPS = 150; POH = 10 + 150 − 40 = 120.
4. Weeks 3, 4: POH = 80, 40.
5. Week 5: 40 − 60 < 0, so MPS = 150; POH = 130. Weeks 6, 7: 70, 10.
6. Week 8: 10 − 60 < 0, so MPS = 150; POH = 100.
7. MPS: 150 units in weeks 2, 5 and 8. POH: 10, 120, 80, 40, 130, 70, 10, 100.
8. ATP week 1 = 60 − 50 = 10. ATP week 2 = 150 − (35 + 20 + 10) = 85. ATP week 5 = 150 − (5 + 0 + 0) = 145. ATP week 8 = 150.
9. ATP: 10 (wk 1), 85 (wk 2), 145 (wk 5), 150 (wk 8).

Example 2 (GATE level) – with safety stock, and an order-promising decision. On-hand = 45 units, lot size = 80 units, safety stock = 10 units. Forecast, weeks 1–6: 30, 30, 30, 30, 40, 40. Customer orders: 33, 20, 10, 4, 2, 0.

1. Requirements: 33, 30, 30, 30, 40, 40.
2. Week 1: POH = 45 − 33 = 12 ≥ 10, no MPS.
3. Week 2: 12 − 30 = −18 < 10, so MPS = 80; POH = 62.
4. Week 3: POH = 62 − 30 = 32.
5. Week 4: 32 − 30 = 2 < 10, so MPS = 80; POH = 82.
6. Week 5: POH = 42. Week 6: 42 − 40 = 2 < 10, so MPS = 80; POH = 82.
7. MPS = 80 units in weeks 2, 4 and 6.
8. ATP: week 1 = 45 − 33 = 12; week 2 = 80 − (20 + 10) = 50; week 4 = 80 − (4 + 2) = 74; week 6 = 80.
9. A customer asks for 60 extra units in week 3. Cumulative ATP available by week 3 = 12 + 50 = 62 ≥ 60, so the order can be promised for week 3; ATP for weeks 1–2 drops to 2 units in total.

Common mistakes

• Using forecast plus orders as the requirement (double counting); take the larger of the two.
• Including forecast in the ATP calculation; ATP subtracts only booked orders.
• Adding on-hand stock to the ATP of later MPS periods; it belongs only in period 1.
• Triggering a lot when POH reaches safety stock exactly; a lot is needed only when POH would fall below it.
• Master-scheduling every end-item variant in an assemble-to-order business instead of the modules.
• Changing quantities inside the frozen zone because "the system allows it".

For GATE PI

Questions give an MPS grid (on-hand, lot size, forecast, orders) and ask for the projected on-hand in a given week, the week of the next MPS receipt or the ATP in a given week. Conceptual items cover time fences, the position of the MPS between aggregate planning and MRP, and what is master-scheduled in MTS/ATO/MTO environments. Practise filling the grid row by row and computing discrete ATP.

Quick check

1. Forecast 70, booked orders 85. What requirement is used for the week?
2. POH at the end of last week was 25; this week requirement is 40, lot size 100, no safety stock. What are MPS and POH this week?
3. On-hand 30, week-1 MPS 0, orders in weeks 1 and 2 are 18 and 7, next MPS receipt in week 3. What is the week-1 ATP?
4. In which time-fence zone can the MPS be changed freely?

Answers: 1. 85 units. 2. MPS = 100, POH = 85 units. 3. 30 − (18 + 7) = 5 units. 4. The liquid zone beyond the planning time fence.