Development of a Fashion Sales Forecasting Decision Support System Using Artificial Intelligence Techniques

Level of sales forecasting accuracy plays a crucial role in determining the profit margins of fashion retailers. In today's retail industry, fashion retailers, including the four sponsoring companies in this project, are still relying on the subjective assessment and experience to make forecasting decisions. This type of experience will not be stored when the professionals leave the companies while inexperienced buyers may not be capable or confident to develop a reliable sales forecast. A forecasting system completely meeting their needs is requested but is absent in the market. A scientific sales forecasting decision support system using artificial intelligence techniques, with a focus on its applications to the expanding Greater China retail market, is proposed. Relevant econometric, demographic and psychographic data, as well as the other parameters influencing the sales forecast will be formulated into the system. The system to be developed will provide a state-of-the-art and user-friendly platform for conducting scientific forecasting for both aggregate yearly fashion demands and seasonal sales pattern of various fashion product categories. Retail companies can make use of the system to perform sales forecasting with a higher degree of accuracy for their decisions on product design, production quantities, planned promotion and marketing schemes.