A performance comparison between hybrid and neighbourhood genetic algorithm applied to assembly line balancing problem

Abstract

The lively field of assembly line balancing often have a significant impact on the performance of production systems. In this context, assembly line balancing problem (ALBP) are widely cited in the literature, and known to be NP-hard in general. For the resolution of such problem, a genetic algorithm (GA) is implemented to solve ALBP. There are two objectives to be achieved: to maximize the line efficiency, and to balance the workload between workstations. The GA may lack the capability of exploring the solution space effectively. To improve the search performance of GA, we have adapted two methods to the classical GA, one is a hybridization of GA which is performed by a priority rule-based procedure aiming at seeding the initial population with good solutions, and other one is GA adopted with strategy of localized evolution. Through computational experiments using ten test problems collected from literature, the performance comparison between GAs is performed and the results are reported.