Otimização de layouts industriais através do método enxame de partículas.

Abstract

The facility layout problem is old and it consists, basically, in finding the best way of arrangement of the facilities (machines, departments, etc.) inside of an application field (industry, hospital, banc, etc.). Although the problem is easy to understand, its solution is hard due to the computational complexity. Different methods were used in its resolution and the search for more efficient methods due to the scientific and economic importance still takes place. This work proposes the use of the method of Particle Swarm Optimization - PSO to optimize the layout problem. Particle Swarm is an evolutionary metaheuristic based in population (swarm), formed by individuals (particles), whose evolution happens through the speed. It is an algorithm of simple concept, easy implementation, robustness to control parameters and computational efficiency during the optimization process. Each particle possesses a speed that allows it to travel the space of solutions and a position in that space. To each iteration, and for each particle, the speed is updated in agreement with the previous speed added to the cognitive part of the formula, that represents the knowledge and to the social part that represents the cooperation among the particles. The new position of the particle is determined by the sum of its current position and the new speed. The present work presents two approaches, developed from the algorithm particles swarm to solve the facility layout problem. The second approach presented better results and compared to other results found in the literature it is possible to prove the efficiency of the method Particle Swarm for the facility layout problem for all the tested instances.