A Roadmap for Managing Forecast Consumption

Nawneet Sinha, Senior Manager, Planning

Forecast consumption is a critical process in releasing demand from demand planning into supply planning. Accurate demand planning helps generate accurate demand forecasts, which form the basis of supply planning to build inventory.

In a make-to-stock planning environment, the demand forecast is used to build-up inventory at the finished goods level, to serve customers as quickly and efficiently as possible. Since the make-to-stock strategy focuses on building inventory against forecast, when sales orders come in, the corresponding forecast is removed.

The most important thing is deciding how consumption should happen, in a way that is closely aligned with demand planning as well as supply planning processes.

However, the challenge is that supply planning occurs at the product-location level, while demand planning occurs at an aggregate level, i.e., at the product-family-customer region level. Hence, forecast consumption at the product-location level may lead to suboptimal and inaccurate planning, as it does not align with the demand planning level.

If forecast consumption is not carried out efficiently, it can undo the potential benefits of better forecast accuracy on the demand planning side and may lead to high inventory, shortages and higher cost of delivery.

Thus, desired deliberations and solutioning are required to build an appropriate forecast consumption logic, along with an associated disaggregation logic, to optimally transition forecast from demand planning into supply planning.

Following is a roadmap to help you determine the best forecast consumption solution for your business:

  • Usually, demand forecasting does not happen at product-location level, so forecast consumption should also not be calculated at the product-location level. Instead, it should be measured at the level at which the forecast accuracy is measured and owned by the demand planning team.
  • Forecast after consumption must be suitably disaggregated to the product-location level for supply planning. If required, separate logic should be built-in for hierarchy disaggregation and time disaggregation.
  • Even though the forecast is disaggregated to the lowest product-location level, the approach should be to allow postponement as much as possible so that the supply planning aligns as closely as possible to the demand planning aggregate level. If possible, inventory at the saleable product-location level should be based on safety stock target and actual sales orders, to enable postponement.
  • Decoupling points need to be determined in the network, with best alignment to forecasting, so that the inventory can be primarily buffered at such nodes against the forecast signals.
  • Short-term sales velocity and other inputs should be effectively measured and used to determine safety stock targets at the lowest saleable product-location levels.
  • Since every company has their own set of guidelines for demand planning and supply planning processes, there can’t be a one-size-fits-all solution that can provide the most optimal forecast consumption for everyone. Hence, before the best solution approach for forecast consumption is finalized, all associated processes on both the demand side and the supply side need to be discussed and accounted for.

Read more about The New Normal in Demand Planning.