Hierarchical sales forecasting system for apparel companies and supply chains

Abstract

The typical problems facing with apparel companies and supply chains are forecasting errors, because fashion markets are volatile and difficult to predict. For that reason, the ability to develop accurate sales forecasts is critical in the industry. There are several research studies related to forecasting apparel goods, but very often only for one level. However, apparel companies and supply chains deal with a number of levels at which the forecasts could exist and require consistent forecasts at all of them. The paper presents a hierarchical middle-term forecasting system designed for this purpose on the basis of a literature review. The system is built by the top-down forecasting approach and verified by means of a case study in a particular apparel company. The weaknesses of the system are identified during discussion of the results acquired. A generalised concept of the ANN forecasting model is designed for elimination these weaknesses.