A Biased-Randomized Algorithm for Redistribution of Perishable Food Inventories in Supermarket Chains

Abstract

In supermarkets, perishable products need to be sold to consumers before a given deadline, after which their monetary value is significantly diminished or even completely lost. In the case of valuable products that should not be wasted, the following operational decision needs to be made as this deadline approaches: which is the best way to reallocate products from stores with surplus inventories to stores with unsatisfied demand? This question results in an optimization problem in which the goal is to minimize total transport cost plus opportunity cost associated with a reduction in market value of products being delivered after a given deadline. Our paper examines this inventory reallocation problem, which is modeled as an extension of the multi-depot vehicle routing problem with soft deadlines, that is, delivery deadlines can be violated by incurring an opportunity or penalty cost. Being an NP-hard optimization problem, a metaheuristic algorithm using biased-randomization techniques is proposed as an effective solution approach. A series of computational experiments contribute to validate our algorithm and to illustrate the potential benefits that can be obtained by reallocating perishable and valuable products in various problem settings.