Data mining-based algorithm for assortment planning

With increasing varieties and products, management of limited shelf space becomes quite difficult for retailers. Hence, an efficient product assortment, which in turn helps to plan the organization of various products across limited shelf space, is extremely important for retailers. Products can be distinguished based on quality, price, brand, and other attributes, and decision needs to be made about an assortment of the products based on these attributes. An efficient assortment planning improves the financial performance of the retailer by increasing profits and reducing operational costs. Clustering techniques can be very effective in grouping products, stores, etc. and help managers solve the problem of assortment planning. This paper proposes data mining approaches for assortment planning for profit maximization with space, and cost constraints by mapping it into well-known knapsack problem.