Algoritmo genético para el problema logístico de asignación de la mochila (Knapsack Problem)

The present work addresses the case study of a national trading company through the Knapsack Problem (KP). This logistic problem consists on determining the products to be stored that have the best cost-benefit ratio for the company without affecting its storage capacity. This is the reason why the KP a binary multi-criterion problem of NP-hard complexity. While there are exact tools that can solve some instances of this problem, in practice these can have a significant cost to the company which may limit their acquisition. Due to this, a free-access proposal which consists of a Genetic Algorithm (GA) was developed to solve the KP in a feasible way for the company. The GA uses random crossing, uniform crossing and mutation operators. The performance of the GA was evaluated with an optimal value, showing a percentage error smaller than 4% when compared to an exact commercial tool.