An Automatic Textile Sales Forecast Using Fuzzy Treatment of Explanatory Variables

To reduce their stocks and to limit ruptures, textile companies must improve their supply chain management. This organization requires sales forecasting systems adapted to the uncertain environment of the textile field. The uncertainty is characterized by noisy data, short historic and numerous explanatory variables that influence the sales behavior. This paper deals with new forecasting models based on "soft computing" and more particularly, last evolutions of hybrid fuzzy model (HFCCX) developed in previous works. HFCCX model uses fuzzy logic abilities to map the non-linear influences of explanatory variables to perform mean-term forecasting. The drawback of this model is the require of an expert judgment for the learning process. The last improvements of our model called AHFCCX allow an automatic learning of the explanatory variables influence. To evaluate performances, a comparative test between AHFCCX, HFCCX and classical models has been applied to real data of textile items selected from an important French ready-to-wear distributor.