An Evolutionary Algorithm for the Product to Shelf Allocation Problem

In this paper we propose an evolutionary algorithm to address the problem of allocating products to shelves in a supermarket (product to shelf allocation problem or P2SAP) and show several instances where it was applied successfully. We first show the main problem exact methods pose, namely bad scalability properties. This means the computational time is of the order of a few minutes for the simplest of cases (one shelf with few modules; few products), while for more complex problems it exceeds 30 hours or, worse still, the method does not provide a solution at all. We then propose an evolutionary algorithm and test it on four different problem configurations (three with one shelf and one with two shelves). In all cases acceptable results can be obtained in a very short timescale.