A Genetic Algorithm to Solve a Multi-Product Distribution Problem

Abstract

The dissertation aims at creating a tool to support the decision making of downstream managers of a portuguese fashion company. The managers must decide whether to send a certain product to each store, and specify the quantities, from one warehouse. We chose to approach the problem with a cost minimization perspective, but with special interest on the constraints.

Being a hard combinatorial problem, a heuristic had to be used. Since Genetic Algorithms have been known to be efficient at solving this kind of problems we propose to build one suited for it. The results will show that the algorithm is capable of solving the problem efficiently, effectively and, ultimately, being a decision support tool.