- Most of what sits in the warehouse goes back to proposals that SAP generates from the parameters in the material master: lot size, safety stock, planned delivery time, MRP type, and reorder point.
- Start with the materials that tie up a lot of capital, segmented by consumption value and demand pattern. Each segment gets its own rules instead of a blanket rule.
- Common causes are lot sizes from old supplier terms and reorder points nobody adjusts. Another is double buffering through safety stock, safety time, and overly long planned delivery times.
- Make changes in waves and measure inventory and availability for each segment. New software pays off if it performs measurably better than well-maintained parameters.
Where inventory comes from
Inventory that is too high is rarely the result of individual wrong decisions. Usually, SAP proposes exactly what the parameters in the material master specify, and planners (MRP controllers) order it. Many parameters were set during implementation and have not been reviewed since, although demand, suppliers, and the product mix have changed.
| Inventory component | How it builds up | Lever in SAP |
|---|---|---|
| Cycle stock | Each order covers more than the next requirement | Lot-sizing procedure, fixed lot size, minimum lot size, rounding value |
| Safety stock | Buffer against fluctuations in demand and lead time | Safety stock, service level, range of coverage profile |
| Time buffer | Safety time and long goods receipt processing times bring goods in before they are needed; overly long planned delivery times trigger earlier orders | Safety time, goods receipt processing time, planned delivery time |
| Excess stock | Forecast or planned independent requirements too high, exception messages left unprocessed | Forecast model, planned independent requirements, MRP list |
| Dead stock | Phase-out parts and spare parts whose demand has disappeared | Discontinuation with a follow-up material; decision on stocking, selling, or scrapping |
Each component has its own cause. That is why you can check and reduce each one separately.
Segment by value and demand pattern first
If you treat all materials alike, effort and inventory get spread thin. So segment your materials first, by consumption value (ABC) and by how regular the demand is (XYZ).
SAP itself provides the data. In SAP ECC, and still in SAP S/4HANA, Inventory Controlling shows the ABC analysis (MC40), slow-moving items (MC46), and dead stock (MC50). S/4HANA adds Fiori apps such as “Slow or Non-Moving Materials” (F2137) and “Dead Stock Analysis” (F2899).
You calculate the XYZ classification from the coefficient of variation of consumption per period, that is, the standard deviation divided by the mean.
| Segment | Typical setting | Question to ask |
|---|---|---|
| AX: high value, regular demand | Deterministic MRP, forecast values as planned independent requirements, small or period lot sizes, tight safety stock | How many weeks of demand does one order add to stock? |
| AZ: high value, intermittent demand | Procure against actual requirements, safety stock based on the consequences of a shortage | What does it cost if the part is missing? |
| CX: low value, regular demand | Reorder point planning, larger lot sizes, little maintenance effort | Is the effort worth it for every single order? |
| CZ: low value, intermittent demand | Small reorder point or no stock at all | Does the part need to be in stock at all? |
For B and Y materials, the rules fall in between. Z materials with intermittent demand are often spare parts; there are dedicated forecasting methods for them. For each segment, you also set a service level, that is, how reliably the material should be available from stock.
To start, one list is enough: the A materials, sorted by days of supply. At the top are the parts that tie up the most capital relative to their consumption. Check lot size, buffers, and planned delivery time there first.
How much inventory each lot size creates
With steady consumption, half the lot size sits in stock on average, plus the safety stock. If you halve the lot size, this cycle stock halves, and the number of orders doubles. Every lot size is therefore a trade-off between ordering cost and tied-up capital.
An example with assumed values: weekly demand of 100 units and a fixed lot size of 2,000 units. On average, 1,000 units then sit in stock, ten weeks of demand. With 400 units per order, it would be 200 units, with five times as many orders.
| Procedure in SAP | How the quantity is set | Good fit for | Watch out for |
|---|---|---|---|
| Lot-for-lot order quantity (EX) | Exactly the shortfall | Expensive materials with predictable demand | Many small orders when requirements are fragmented |
| Fixed lot size (FX) | Always the same quantity, multiples if needed | Fixed container, transport, or production quantities | Quantities from old supplier terms that nobody asks for today |
| Period lot size (TB, WB, MB) | All requirements of a day, week, or month in one receipt | Regular demand, fixed delivery days | A monthly lot can add up to a month of demand to stock |
| Replenishment up to maximum stock level (HB) | Stock is replenished up to a fixed upper limit | Reorder point planning, limited storage space | A maximum stock level unrelated to today’s consumption |
| Optimum lot-sizing procedures, such as part period balancing, least unit cost procedure, or Groff reorder procedure | Trade-off between lot-size-independent costs and storage costs | Materials with significant ordering or setup costs | They calculate with the cost rates in the system, even if these are only estimates |
Minimum lot size and rounding value come on top. Rounding to full pallets turns a small shortfall into a pallet of inventory. Check whether the rounding fits consumption or only the supplier’s packaging.
Safety stock needs a reason
Safety stock is meant to absorb fluctuations in demand and lead time, not to cover planned demand. If it is never touched, it is probably too high. If stock keeps falling below it, the planning upstream is usually off, for example the forecast or the planned delivery time.
SAP offers several ways to set it.
- Fixed safety stock in the material master. It is easy to maintain but stays put when consumption drops.
- Calculation by SAP in automatic reorder point planning. With every forecast run, SAP recalculates safety stock and reorder point from service level, replenishment lead time, and the forecast’s mean absolute deviation.
- Dynamic safety stock through a range of coverage profile. It is defined in days of supply and grows and shrinks with average daily requirements.
- Safety time. It moves receipts forward by a few days. The buffer then appears in no stock field, but it still sits in the warehouse.
It is the combination that gets expensive. Safety stock, safety time, and a generous planned delivery time on the same material buffer the same uncertainty three times.
If a central warehouse supplies several plants, a second stocking level comes into play. If the plant and the central warehouse both stock the same part, both buffer the same uncertainty. We are currently developing materials planning concepts for such a central warehouse in an ongoing engagement.
Do MRP type and reorder point fit the demand?
The MRP type in the material master determines which procedure SAP uses to plan a material. Deterministic MRP plans against concrete requirements from sales orders, bills of material, reservations, and planned independent requirements. Consumption-based procedures infer future consumption from past consumption.
| MRP type | Procedure | Good fit for | Typical mistake |
|---|---|---|---|
| PD | Deterministic MRP | Materials with known requirements from orders, production, or maintenance | Planned independent requirements that nobody maintains create receipts without real demand |
| VB, VM | Reorder point planning, manual or automatic | Regular consumption, mostly low-value parts | Reorder point set by hand and unchanged for years |
| VV | Forecast-based planning | Regular, predictable consumption; in S/4HANA only in the compatibility scope | The forecast model no longer fits, and nobody measures forecast accuracy |
The reorder point should cover demand during the replenishment lead time, plus the safety stock. If it is higher, you order too early. A reorder point set by hand becomes outdated as soon as consumption declines.
For planners in live operations on S/4HANA, we designed a reorder point report.
Planned delivery time and exception messages
The planned delivery time determines how far ahead of the requirement SAP proposes an order. If it is too long, you order earlier than necessary; in automatic reorder point planning, reorder point and safety stock also rise. If it is too short, materials are missing, and planners have to reorder with a rush surcharge.
Compare the planned delivery times in material masters and purchasing info records with actual delivery times from purchase orders. Define who maintains them and how often.
Good parameters help little if nobody reacts to deviations. When demand drops, SAP uses exception messages to propose postponing or canceling open receipts of materials planned with deterministic MRP. If these messages sit unprocessed in a long MRP list, the goods arrive anyway.
So filter the MRP list for the exceptions that change inventory or availability, and work through those first. For inventory, messages 15 and 20 matter most: with them, SAP proposes rescheduling receipts out or canceling them.
When new software pays off
Planning and optimization tools can do more than standard MRP, for example balance inventory across several stocking levels or calculate probabilistic forecasts. But they only show their value on well-maintained master data. With outdated planned delivery times and blanket lot sizes, every tool calculates with wrong assumptions.
Before selecting a tool, define how you will measure success. In a test, compare the tool with your well-maintained SAP parameters, on the same history and with the same KPIs for inventory and availability.
How to check your parameters
These steps need no new tool, only access to your system and time from materials planning and purchasing.
- Record the baseline. Inventory value, days of supply, and shortages per plant and material group, so that you can prove the effect later.
- Segment the materials. ABC by consumption value, XYZ by regularity of demand, plus slow-moving items and dead stock.
- Check lot sizes. Compare procedures, fixed lot sizes, minimum lot sizes, and rounding values of the A materials with today’s demand.
- Look for double buffers. Materials that carry safety stock, safety time, and an overly long planned delivery time at once. A table extract from MARC with safety stock (EISBE), safety time (SHFLG, SHZET), and planned delivery time (PLIFZ) shows them at a glance.
- Compare planned delivery times. Planned against actual from the purchase orders of recent months, with a rule for maintenance.
- Set the procedure for each segment. MRP type, reorder point, and service level to match the demand pattern.
- Sort exception messages. Filter the MRP list so that reschedulings and cancellations are processed first.
- Change in waves. One material group at a time, documented and using mass maintenance (MM17), with a measurement before the next wave.
Frequently asked questions
How quickly does inventory drop after a change?
It starts with the next order, which is slower than many expect. Stock on hand and on order only goes down through consumption, cancellation, or stock transfer. So also check open purchase orders that the new parameters would no longer require.
Stock that will no longer be consumed has to be reduced through sale, return, or scrapping. This can weigh on earnings, so agree on the reduction with finance in advance.
Can’t we just cut safety stocks across the board?
A blanket cut hits fast-moving and critical parts just as hard as slow-moving items. Inventory drops, and shortages rise. Cut by segment and by cause.
Do we need an add-on tool for inventory optimization in SAP?
Not at first. Well-maintained parameters, segmentation by value and demand pattern, and an MRP list that shows exceptions are the foundation for any tool. A test with your data shows whether a tool delivers more.
Should the parameters be part of the S/4HANA transition?
Yes. In a new implementation, you migrate the material masters anyway; that is a good moment to clean up parameters during data migration. In a system conversion, old values carry over, so check them before the conversion or in the first weeks after it.
Pay attention to the MRP type VV. In S/4HANA, forecast-based planning is part of the compatibility scope, whose usage rights generally ended at the end of 2025. The final transition period expired at the end of May 2026.
MRP Live, the planning run in S/4HANA, plans these materials only with the classic MRP logic. SAP recommends transferring the forecast values as planned independent requirements and planning the materials with deterministic MRP.
How we help
If daily business leaves your planners no time for these steps, we get things started with an analysis of inventory and MRP parameters. You get findings in order of priority and proposals for parameters by material segment that your planners can put into practice.
We have designed and implemented materials planning procedures; on request, we make the changes in the system ourselves. If you are choosing planning software, we advise you independently of any software vendor, with your well-maintained parameters as the benchmark. All of this is part of our supply chain planning and inventory optimization service.