- Supply Chain Management
- Processes
- Demand planning
SAP-based solutions for demand planners
Central questions in demand planning
- Which products are relevant for planning?
- Which forecasting methods fit best?
- How good was your planning? Target/actual situation comparison:
- How can different futures be simulated?
Knowing today what will be needed tomorrow for reliable demand planning
Reliable sales forecasts are at the heart of the supply chain. Material requirements and machine capacities can only be planned in good time with an accurate assessment of future sales volumes.
If this is done on a rolling basis, the planner becomes increasingly certain over time about the real situation in terms of material and machinery.
A simulation enables the assessment of worst and best-case scenarios. This makes it easy to simulate external and internal influences on demand planning.
Taking all relevant information into account
It is important to include as much information as possible that has an impact.
This includes not only valid information from marketing and sales, but also information about events, customer assessments, management guidelines and much more.
Clever and accurate demand planning has a direct impact on the subsequent steps in the supply chain process namely manufacturing, procurement and inventory management.
For optimal demand planning, data from the past must be interpreted correctly to be able to make valid forecasts for the future. The ifm software uses various statistical methods for this purpose. The illustrations show that not every forecasting method optimally extrapolates reality.
Croston forecasting method
Seasonal forecasting method
The figures show two different forecasting methods that are intended to extrapolate the actual development. Although the "seasonal method" shows a more realistic continuation than the Croston forecast, neither forecast model is ideal for providing valid information on sales trends. The actual value of the past (shown as blue columns) deviates significantly from the green forecast columns in both methods.
Prophet is the winner
This illustration shows the additive model, better known as the Prophet method. Forecast and actual values are almost congruent. The Prophet method emerges as the clear winner: the historical data aligns convincingly well and suggests a reliable extrapolation with a high degree of certainty.
Forecast method: Additive Model / Prophet
Identify suitable forecasting models at first glance!
This figure shows how differently the respective forecasting methods pick up and extrapolate the stock values. While Croston (orange dashed line) fluctuates around the mean value, Seasonal (purple dashed line) shows deflections that are significantly above or below the actual values. The green area on the right-hand side of the image then shows the extrapolations of the various forecast models and the trend line (dashed red). The variety of forecasting models used and the automatic determination of the most suitable one is a key strength of the ifm solution.
Whether in line or bar charts, ifm's forecasting software offers various visualisation options so that the user can compare the fit of the forecasting models at a glance. "See and understand" - the user usually recognises the winner at first glance.
Comparison of forecasting methods
A lot of information helps a lot
The solution takes into account different sources such as marketing, sales or even from the customer side itself to create a comprehensive picture.
1 Marketing ⋅ 2 Sales ⋅ 3 Customer
One truth for all
The database is always up to date and can be viewed and supplemented with information by all participating departments – this means the end of Excelitis and interfaces.
1 Work preparation⋅ 2 Scheduling ⋅ 3 Production ⋅ 4 Quality ⋅ 5 Logistics
More graphics – simpler interpretation
As pictures say more than a thousand numbers, the ifm software even presents complex facts such as the forecasting procedures graphically. They can, therefore, be identified at a glance.
Planning without control is pointless
The forecasting process must be constantly scrutinised to make optimal forecasts in the long-term:
- Were the correct forecasting methods chosen?
- Have the correct sources been integrated?
- Has the information been weighted correctly?
- Does the planning reflect the reality?
After all, demand planning should function according to the principle Planning without control is pointless. Control without planning is impossible.
Only those who recognise their mistakes can improve – by the way, this applies just as much to humans as it does to artificial intelligence.
Not every material needs to be planned, but the right one does.
Even at 80 % planned materials, demand planning can deliver disastrous results. This is because if the 20% of unplanned materials, of all things, account for the majority, planning becomes a product of chance.
So how can the relevant materials be determined?
Simulating future scenarios
Good planners often already have different versions of the future in mind, since the occurrence or non-occurrence of certain events results in different potential scenarios.
In that respect, it is advantageous to plan the different scenarios in advance and calculate the respective effects on the sales plan.
This way you always have the famous Plan B to hand.
Request a demonstration
Would you like to learn more about our supply chain optimisation solutions in SAP? Are you looking for a solution that allows you to automatically transfer data from machines or sensors to SAP and start intelligent processes? Complete the form and request your demonstration on-site or online now.