Forecast accuracy

Demand planning

Demand forecasting

Short term demand forecasting

An important question for demand planning in all businesses is whether all or a selection of products should be sourced for stock. This matter should be resolved by analysis taking into account the risk of lost sales, frequency of demand, sales volumes, cost of inventory and the achievable forecast accuracy.

Once it has been decided to hold inventory, forecast accuracy over a horizon of six to twelve months becomes critical, because the variation in forecast error is a major factor in the calculation of safety stock.

Methods for short term demand forecasting

For short term demand planning it is often practical to assume that external factors such as the prevailing competitive structure and economic background will stay relatively constant. Therefore demand forecasting techniques tend to major on time series projection, the forecast being based on a historical analysis of trend and seasonality. Whereas curve fitting methods may be useful for medium to long term demand forecasting, in the short term a simple approach is usually adequate. Exponentially weighted moving averages are particularly useful, as employed in the family of techniques known as exponential smoothing.

Forecast accuracy and commercial judgement

To maximise forecast accuracy it is always valid to apply market intelligence to the forecast to take account of known events. The process of obtaining sales force collaboration in the demand forecasting and demand planning process needs to be carefully resolved, with appropriate management of multiple versions of the forecast leading to a final 'one forecast' solution for the whole business.

Forecast Solutions can help by carrying out a forecast accuracy health check or by creating a set of forecasts for evaluative purposes or as a sales forecasting service. Advice and training on demand planning methods and procedures can be provided, including tailored demand forecasting training courses.