Lesson 6 of 6
Testing the production order lifecycle
9 min
Listen to this lesson
Production orders touch planning, inventory, costing and finance in one object. That makes them ideal for end-to-end tests and dangerous to test only on the happy path. For Contoso's system conversion, Priya's team re-runs the production order lifecycle in CMQ after each mock conversion and compares results with the same scenarios run in the legacy ECC system EQ1.
Scenarios worth testing
| Scenario | What to prove | Evidence |
|---|---|---|
| Happy path, 200 pc | Each step posts once; statuses move CRTD → REL → CNF → DLV → TECO | Status history, material documents, journal entries |
| Partial confirmations | PCNF after the first, CNF after the last; quantities add up | Confirmation list per operation |
| Scrap at assembly | Scrap is recorded with a reason; receipt is 195; variance includes scrap cost | Confirmation, goods receipt, order cost report |
| Backflush with shortage | Confirmation saves; failed issue appears in COGI; reprocessing clears it | COGI before and after, material document from reprocessing |
| Release with missing parts | Configured behavior (block or user decides) happens | Missing-parts log, status MSPT |
| Wrong confirmation | Cancellation reverses activities and backflushed components | Cancelled confirmation, reversal documents (262) |
| TECO with open quantity | Reservations and capacity disappear; late posting still possible | Stock/requirements list before and after; late document |
A minimal production order regression pack.
Evidence that convinces finance and auditors
For each scenario, capture the order number, the status history, every material document number with movement type, and the journal entries they created. Then reconcile: component issues at standard price plus activity allocations should equal the order's actual debit; the goods receipt should equal yield × standard price. A screenshot of a green message is not evidence; document numbers and amounts are.
- BeforeStock of each component in 171L and of CMC-ALT-100 in 171B; COGI empty for plant 1710
- ExpectedComponents down by BOM × yield (+ scrap); finished goods up by 195; variance = actual − 195 × $150
- ExecuteRun the scenario with named test data and record each document number
- AfterSame stock queries, order cost report, COGI list
- CompareDifferences explained or logged as defects with document numbers
Two conversion-specific risks deserve their own tests. First, custom code: Kenji's team should list any user exits or enhancements in order release, confirmation and goods movements and confirm they still run after the conversion. Second, open orders: Luis's cutover plan must say which orders are in process at cutover, and the first-week tests in production must confirm that orders converted from ECC can still be confirmed, received and set to TECO.
What would you do?
In the second mock conversion, every production order test passes in CMQ, but nobody ran a backflush with a shortage. Priya has to decide whether to sign off the manufacturing test cycle. What should she do?
After go-live, the same pack becomes the production order part of release regression in CMR. The Test Automation Tool for SAP S/4HANA Cloud Public Edition and SAP Cloud ALM test automation serve the same purpose in their landscapes: the scripts carry the expected values, and every run compares document chains, not just screens.
- Test the lifecycle end to end plus the exceptions: partial confirmation, scrap, backflush shortage, cancellation, TECO.
- Evidence is document numbers, statuses and journal entries, reconciled to order costs.
- Use a before–expected–after pattern on stock and COGI.
- In a conversion, test custom enhancements and orders open at cutover.
Check your understanding · not graded
Lesson check
1. Which evidence best proves a backflush shortage was handled?
2. In a 200-piece order with 5 scrapped, what should the goods receipt quantity be in the test?
3. Why test orders that are open at cutover?
4. Which reconciliation proves an order's costs are complete?