Applying multi-objective ant colony optimization algorithm for solving the unequal area facility layout problems

• Published in: Applied soft computing Vol. 74; pp. 167 - 189
• Main Authors: Liu, Jingfa, Liu, Jun
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.01.2019
• Subjects: Ant colony optimization; Facility layout problem; Multi-objective optimization; Pareto optimal; Preference; Facility layout problem; Ant colony optimization; Multi-objective optimization; Pareto optimal; Preference
• ISSN: 1568-4946; 1872-9681
• DOI: 10.1016/j.asoc.2018.10.012

The unequal area facility layout problem (UA-FLP) which deals with the layout of departments in a facility comprises of a class of extremely difficult and widely applicable multi-objective optimization problems with constraints arising in diverse areas and meeting the requirements for real-world applications. Based on the heuristic strategy, the problem is first converted into an unconstrained optimization problem. Then, we use a modified version of the multi-objective ant colony optimization (MOACO) algorithm which is a heuristic global optimization algorithm and has shown promising performances in solving many optimization problems to solve the multi-objective UA-FLP. In the modified MOACO algorithm, the ACO with heuristic layout updating strategy which is proposed to update the layouts and add the diversity of solutions is a discrete ACO algorithm, with a difference from general ACO algorithms for discrete domains which perform an incremental construction of solutions but the ACO in this paper does not. We propose a novel pheromone update method and combine the Pareto optimization based on the local pheromone communication and the global search based on the niche technology to obtain Pareto-optimal solutions of the problem. In addition, the combination of the local search based on the adaptive gradient method and the heuristic department deformation strategy is applied to deal with the non-overlapping constraint between departments so as to obtain feasible solutions. Ten benchmark instances from the literature are tested. The experimental results show that the proposed MOACO algorithm is an effective method for solving the UA-FLP. •A multi-objective ACO algorithm is proposed to solve the UA-FLP.•A heuristic strategy is used to convert the constrained problem into an unconstrained one.•A local search and department deformation are applied to obtain feasible solutions.•Pareto optimization and niche technology are used to obtain Pareto-optimal solutions.