Manufacturing Intelligence Series | Module 5 of 6

Demand planning:
Less stock, fewer rush orders

Over- and underproduction are two sides of the same forecast gap. Forecasting on your ERP and PPS history reduces tied-up stock and avoids the costly rush orders that patch the gap today.

01 The problem – Too much ties up capital, too little costs rush surcharges

Poor demand forecasts make both directions expensive

Overproduction ties up capital, blocks storage space and partly ends in scrapping. Underproduction triggers rush orders, special transports and extra shifts. Both stem from the same cause: a demand forecast based on gut feeling and last year.

Your history holds the patterns: seasonality, order lead time, call-off behaviour per customer, correlations between items. Forecasting turns this into a reliable demand curve per item and week.

−15% stock + fewer rush costs = €294,000 / year

Less tied-up capital, without endangering delivery capability.

Order history
Call-off patterns
Forecasting
Demand curve
Smoothed planning

02 The model – Demand per item and week

Gradient boosting and time series on order, call-off and seasonal patterns

▸ Output
WAPE Bedarfsprognose: 14.2%
Ø Sicherheitsbestand: 1.040 Stk (vorher pauschal: 1.800 Stk)
Demand forecast with confidence band vs. actual call-off
↳ Forecast with uncertainty band

The model delivers not just a point value but a confidence band. This lets safety stock be controlled per item: high for volatile parts, lean for stable ones. Flat safety stocks across all items – the most common source of excess stock – disappear.

03 Business impact – Cut tied-up capital and rush cost

The lever: lower stock × cost of capital, plus fewer rush and scrapping costs

€294,000
Savings / year
−15%
Less stock
−30%
Fewer rush orders
Savings by lever – capital, rush cost, scrapping
Model calculation · How the figure is built – derived transparently

No round number: every assumption comes from the sample plant and is stored centrally. With your real figures only the input changes, not the method.

ItemValue
Tied-up inventory€9,500,000
Stock reduction 15% × cost of capital 8%€114,000
Rush cost / year → −30%€114,000
Scrapping / year → −30%€66,000
Result: savings / year€294,000

Assumptions of a sample plant – in a real project your data replaces these values.

↳ Three levers from one forecast

The better demand curve works threefold: it lowers tied-up stock (cost of capital), reduces expensive rush orders on shortfall and avoids scrapping on surplus. All three draw from the same model.

04 Next steps

From forecast to integrated planning

① Tap the history

We use order, call-off and stock history from your ERP/PPS – without new data capture.

② Forecast pilot

For one item group the forecast model is built. You see the hit rate validated against recent quarters.

③ Planning proposal

Weekly demand proposal per item with a safety band – as input for your production and procurement planning.