John’s key points are these. • Organizations should consider which service level measure aligns most closely to their corporate objectives. Cost of Forecast Error (CFE) calculations should be based on this measure. • Assessment of service level targets at higher levels of aggregation...

Forecasting seasonal products can be difficult when the products are fairly new or highly variable. The Spring 2007 issue of Foresight contained a special feature on modeling seasonality in short time series. The articles addressed the issues of minimum sample size requirements and surveyed the...

John and Aris distinguish between forecase-accuracy metrics, which measure the errors resulting from a forecast method, and accuracy-implication metrics, which measure the achievement of the organization's stockholding and service-level goals. Both measurements are important. The correct choice...

Slow items with intermittent and lumpy demand patterns can make up a substantial part of an organization's inventory. They are difficult to forecast and some of the most popular forecasting methods are unsuitable. The author describes the principal methods for forecasting intermittent and lumpy...

The idea of forecastability is valuable because it allows attention to shift from forecasting methods to the series that are being forecast, but the concept of forecastability needs sharpening. • Coefficients of variation and approximate entropy metrics assess the stability of a time series....

With this issue, Foresight debuts a new column on supply-chain forecasting. John Boylan, our column editor, will offer his insights on the practical challenges of forecasting and reconciling forecasts across the hierarchical levels of a supply chain. Copyright International Institute of...

Foresight’s Summer 2010 issue contained a letter to the editor from David Hawitt, suggesting that forecast managers would be more receptive to hearing about forecast accuracy rather than forecast error. If forecast error is measured by the average absolute percentage error (MAPE), then...

ABC classifications are commonly used to streamline a large number of SKUs into fewer, more manageable categories. As Aris, John, and Ruud explain, this classification may be useful for inventory control, but it does not provide much help in the selection of appropriate forecasting methods. The...