A solution based on data mining to shelf space allocation

For the retailing, the shelf space allocation is a key issue which impacts on the retailer's product sales and profits directly. For the small-scale supermarket, due to the small store and the product variety, the traditional “format style” shelf space layout is not appropriate. To solve this problem, a shelf space allocation solution based on data mining techniques is proposed in this paper. With the introduction of association distance, we realize the description of the relevance between two any items purchased together in transaction database. With the concept of promotions coefficient, we get the description of the promotions ability of two items. Based on these concepts, with the cluster analysis techniques and the association rule mining technique, we propose a resolution based on “island style”, and get the optimization and automation of shelf space allocation in the small-scale supermarket.