Retailer layout design: a novel hybrid approach with association rules mining and MCRAFT

Spatial layout of a retail store is a crucial decision variable related to both utilisation of store area and purchasing behaviour of the customer. In this respect, the task of optimising the allocation of shelves to specific product segments has become a strategic decision to facilitate a more comfortable shopping environment for customers, which, in turn, increases sales volume. This study proposes a novel hybrid approach to facility layout design problem, which combines association rules mining (ARM) and a facility layout method, MCRAFT. The proposed methodology is composed of two main stages: 1) rule mining; 2) layout design. More specifically, the presented approach exploits the association rules obtained from purchasing records and uses them as a proximity measure input to MCRAFT algorithm to determine the layout of the store. The merit of the proposed methodology is shown with a case study on a prominent Turkish supermarket chain.