Contents
A master production schedule (MPS) states the quantities and completion dates planned for selected finished products or major assemblies. Oracle's MRP glossary calls it "the anticipated build schedule" in rates or discrete quantities and dates. The MPS feeds material planning and the promises sales makes to customers. It does not sequence every machine operation.
The University of Cambridge Institute for Manufacturing describes the MPS as the starting point of MRP II: bills of materials translate the scheduled products into component requirements. The APICS CPIM learning material adds that the master schedules must match the totals of the agreed production plan, so that sales can quote delivery dates from it.
Terms Used in Master Scheduling
| Term | Meaning |
|---|---|
| Gross requirements | Total independent and dependent demand before netting stock and scheduled receipts |
| Scheduled receipt | A released job or purchase order that already counts as supply |
| Available inventory | On-hand stock minus allocations, reservations, backorders and usually quantities held for quality problems |
| Projected available balance | Stock projected forward: usable on-hand plus scheduled receipts plus planned orders, minus gross requirements |
| Available to promise (ATP) | The uncommitted part of inventory and planned production that sales can still promise |
| Time fence | A point in the planning horizon from which restrictions on changes apply |
Available inventory and ATP follow the APICS definitions; the other terms follow Oracle's glossary. Software products differ in details, for example whether planned orders count in the projected balance, so check your system's definition before comparing figures between plants.
State the Timing Assumption
A weekly quantity can hide a delivery problem inside the week. If demand is due on Monday and production finishes on Friday, the weekly balance looks correct while the customer is late. Choose time buckets that match the actual decision.
Also separate on-hand stock from available inventory. Material awaiting inspection or reserved for another commitment cannot cover the demand being planned. The opening balance in the example below is therefore available inventory.
Work Through a Stock Balance
An assembly starts week 1 with 70 available units. The plan keeps 30 units at the end of each week. Demand and receipts count at the end of the week, and each unit needs 0.4 hours on a shared test rig with 60 usable hours per week.
| Week | Opening Available | Demand | MPS Receipt | Closing (Projected Available Balance) | Test Hours Needed |
|---|---|---|---|---|---|
| 1 | 70 | 140 | 100 | 30 | 40 |
| 2 | 30 | 180 | 180 | 30 | 72 |
| 3 | 30 | 110 | 110 | 30 | 44 |
Closing stock equals opening stock plus the MPS receipt minus demand: 70 + 100 − 140 = 30 units in week 1. Week 2 starts with those 30 units, so its 180-unit demand needs a receipt of 180 to keep the same closing stock. The hours column is the receipt times 0.4 hours: 40, 72 and 44.
Check the Limiting Resource
Week 2 needs 72 test hours against 60 available, an overload of 12 hours before any other product uses the rig. This conversion of the master schedule into the load on a key resource is what the Oracle glossary calls rough-cut capacity planning. The quantity balance can be correct while the plan is still not feasible.
Variant: build 30 units of week 2's demand one week early. Week 1 then receives 130 units and needs 52 hours; its closing stock rises to 60. Week 2 receives 150 units and needs exactly 60 hours, and closes at 60 + 150 − 180 = 30. The overload disappears, at the price of holding 30 extra units for a week and of components arriving one week earlier.
Promise Orders From ATP
Available to promise is the part of an MPS receipt that is not yet committed to customer orders. With the discrete method in the APICS excerpt, ATP for a week with a receipt is that receipt minus the customer orders due before the next receipt. Suppose 150 of week 2's 180 units are already booked as orders and the rest is forecast. Then ATP for week 2 is 180 − 150 = 30 units, and sales can promise those 30 without touching another customer's delivery.
Time Fences Need a Reason
Most planning systems call the frozen horizon a time fence. In Oracle's time fence control, the demand time fence marks the period in which planning uses only actual customer orders and ignores the forecast; inside the planning time fence, the planning run does not create new orders or move existing ones earlier for discretely planned items. APICS notes that even lean plants keep time fences for items with long lead times, because a stable schedule supports stable production.
The useful horizon depends on the lead times of your own parts; there is no universal number of frozen weeks. Record what changed and which requirements it affects. Moving a quantity earlier can create a shortage even when the monthly total is unchanged.
Close the Plan With Actual Output
Compare scheduled receipts with the quantities that actually become available. Started units, finished units and accepted units are different states, and a report should keep them apart rather than add them into one production figure.
When receipts repeatedly fall short, check the assumptions behind the limiting operation: processing time, product mix, interruptions or missing material. Moving every missed unit into the next week without revising those assumptions only transfers the overload. SYMESTIC, a cloud MES, records produced quantities and rejects per machine and order automatically, which gives you the completed units to set against each MPS receipt. How the schedule then drives material and capacity planning is covered under MRP II. The 0.4 hours per unit come from the process plan.
Sources
- Oracle Master Scheduling/MRP: Glossary
- APICS CPIM 2019 excerpt (ASCM): Master Planning, Master Scheduling and Sales
- Oracle Master Scheduling/MRP: Time Fence Control
- University of Cambridge IfM: MRP II
See how many units each line actually completes per week, the figure to compare with your MPS receipts, in a free 30-minute live demo using your own application.
See SYMESTIC on your own use case
In a free live demo, a product expert walks you through the running platform and shows what OEE, downtime and the connection of your machines and ERP look like in practice.
- Live OEE
- Downtime and reasons
- Connecting machines and ERP
