Competition and globalization imply a very accurate production of the Textile-Apparel-Distribution network actors. A sales forecasting system is required to response to the versatile textile maeket and the need of the distributor. Nowadays, the existing forecasting models are generally unsuitable to the textile industry. We propose a forecasting system, which is composed of two models and periorm forecasts on mean-term horizon for families and items sales. This system is based on classification procedures, authorizing the data aggregation. Performances of our models are then evalualed using the real data from a important French textile distributor.
sales forecasting, textile distribution, classification.