Optimization of a multi-constraint two-sided assembly line balancing problem using an improved imperialist competitive algorithm

Abstract

PurposeAssembly is the last step in manufacturing processes. The two-sided assembly line balancing problem (TALBP) is a typical research focus in the field of combinatorial optimization. This paper aims to study a multi-constraint TALBP-I (MC-TALBP-I) that involves positional constraints, zoning constraints and synchronism constraints to make TALBP more in line with real production. For enhancing quality of assembly solution, an improved imperialist competitive algorithm (ICA) is designed for solving the problem.Design/methodology/approachA mathematical model for minimizing the weighted sum of the number of mated-stations and stations is established. An improved ICA is designed based on a priority value encoding structure for solving MC-TALBP-I.FindingsThe proposed ICA was tested by several benchmarks involving positional constraints, zoning constraints and synchronism constraints. This algorithm was compared with the late acceptance hill-climbing (LAHC) algorithm in several instances. The results demonstrated that the ICA provides much better performance than the LAHC algorithm.Practical implicationsThe best solution obtained by solving MC-TALBP-I is more feasible for determining the real assembly solution than the best solution obtained by solving based TALBP-I only.Originality/valueA novel ICA based on priority value encoding is proposed in this paper. Initial countries are generated by a heuristic method. An imperialist development strategy is designed to improve the qualities of countries. The effectiveness of the ICA is indicated through a set of benchmarks.