A Methodology Based on Genetic Algorithms and Piecewise Linear Programming for Facility Layout Optimization

Abstract

This paper proposes a methodology that integrates a genetic algorithm and a piecewise linear programming model to optimize facility layout design. This class of problems represents a major challenge in engineering due to its combinatorial and nonlinear nature. The problem considers as input data the areas of the facilities, the material flow between them, and the dimensions of the plant. The main objective is to minimize the total weighted distance between the centroids of the facilities. The genetic algorithm determines the relative arrangement of the facilities, while the piecewise linear programming model accurately adjusts the final dimensions and coordinates of each facility. The results obtained from instances of different sizes demonstrate the effectiveness of the proposed methodology in finding feasible solutions within reasonable computational times.