A hybrid genetic algorithm approach for solving an extension of assembly line balancing problem

Abstract

The lively field of assembly line configuration and adjustment often have a significant impact on the performance of manufacturing systems. In this context, assembly line balancing problems (ALBPs) are widely cited in the literature. An ALBP consists of distributing the total product manufacturing workload among the stations along the manufacturing line. Previous research has focused on developing effective and fast solution methods for solving simple assembly line balancing problems (SALBP) and their various extensions. Each extension is motivated by several real-life applications and the need for solving precise practical problems. In this article, another interesting extension of SALBP (named in this work 'Task Restrictions Assembly Line Balancing Problem' of type 2 (TRALBP-2)) is focused on. In this situation, the number of stations is known and the objective is to minimise cycle time where both precedence and zoning constraints between tasks must be satisfied. For the resolution of such problem, an innovative hybrid genetic algorithm (HGA) scheme hybridised with a local search procedure is implemented. This genetic algorithm consists of a new representation scheme and a special genetic operator. The effectiveness of the proposed HGA is evaluated through various sets of instances which are (1) theoretically and randomly generated, (2) collected from the literature and (3) based on a real case study of an automotive cable manufacturer. Comparison of the proposed HGA results with CPLEX software for the TRALBP-2 demonstrates that, in a reasonable time, the proposed HGA generates consistent solutions that are very close to their optimal ones. Therefore, the proposed HGA approach is very effective and competitive.

An integrated approach for line balancing and AGV scheduling towards smart assembly systems

“…A wealth of literature is available for traditional assembly line balancing problems, and most of these contributions demonstrate the application of optimization techniques for evaluating them (Triki et al, 2016). With respect to robotic assembly lines, Rubinovitz and Bukchin (1991) were the first to propose balancing a robotic assembly line by minimizing the number of workstations.…”

Hybrid genetic algorithm for a type-II robust mixed-model assembly line balancing problem with interval task times

“…The pure genetic algorithm is good at global search but slow to converge [27]. Local search is a promising approach to improve the quality of the objective value and convergence speed [28]. Within the proposed algorithm, a local search procedure is applied to every chromosome of the population.…”

Reconfigurable assembly line balancing for cloud manufacturing

“…For the optimization algorithm, it is obvious that good parameter values are essential for good performance. In this paper, we refer to the parameter set in literature (Delice et al 2016(Delice et al , 2017Triki et al 2016;Quyen et al 2016;Xu et al 2016) and use an automatic configuration method called iterated F-Race (López-Ibáñez et al 2016;Bartz-Beielstein et al 2010;López-Ibáñez and Stützle 2012) to set the parameters. The offline configuration tool I/F-Race we used was provided by the irace package (López-Ibáñez et al 2016) which applied Friedman post-tests to generate new candidate configurations and perform races to discard the worst-performing ones.…”

“…Li et al (2017) put forward a discrete cuckoo search algorithms for two-sided robotic ALBP (Çil et al 2016). An improved GA was introduced by Delice et al (2016), Triki et al (2016 and Quyen et al (2016). Compared with the traditional GA, these methods have higher optimization performance.…”