Bees algorithm for effective supply chains configuration

Abstract

This article proposes a bees algorithm for solving a mathematical model of strategic supply chains planning and design in the globalization context. The model, taken from the literature, involves a multi-echelon, multi-facility and multi-product supply chain in a dynamic environment. Our bees algorithm, the first to be applied to this kind of problem, was adapted specifically to the optimization problem, and its performance was benchmarked against the genetic algorithm (GA) provided by commercial optimization software MATLAB R2011b. This is in order to test its effectiveness in an already known supply chain design problem, which has not been solved yet in optimized way. Several configurations of the algorithm were tested and analysed; the bees algorithm performs better of 10% in quality of solution respect to the GA with equivalent calculation time. The better quality of their solutions and the good calculation time performances have encouraged towards more accurate studies on tuning parameters and an extended utilization of the methodology on problems of logistic systems configuration and supply chain design.