An iterated greedy heuristic for a market segmentation problem with multiple attributes

1 This paper addresses a real-world customer segmentation problem from a beverage distribution 2 firm. The firm wants to partition a set of customers, who share geographical and marketing attributes, 3 into segments according to certain requirements: (a) customers allocated to the same segment must 4 have very similar attributes: type of contract, type of store and the average difference of purchase 5 volume; and (b) compact segments are desired. The main reason for creating a partition with these 6 features is because the firm wants to try different product marketing strategies. In this paper, we 7 propose a detailed attribute formulation and an iterated greedy heuristic that iteratively destroys 8 and reconstructs a given partition. The initial partition is obtained by using a modified k-means 9 algorithm that involves a GRASP philosophy to get the initial configuration of centroids. The 10 heuristic includes an improvement method that employs a variable neighborhood search procedure. 11 Computational results and statistical analyses show the effectiveness of the proposed approach. 12