Abstract
In literature most of the studies on assembly line considered constant task time and minimised cycle time. However, due to several uncertain events in the real environment, task time and flow time (completion time) of tasks varies and it can exceed the cycle time. Moreover, these variations in task time and flow time of tasks are also transferred to the next station and can affect the flow time on next stations. To solve this issue, a multi-objective assembly line balancing problem, aimed to minimise cycle time and maximise the sum of average probability of stations and the probability of the whole assembly line to ensure that the flow time of tasks on different stations will not exceed the cycle time in the presence of the transferred, added or absorbed variations of task times between the stations, is presented. A hybrid Pareto artificial bee colony (HPABC) algorithm is proposed to solve the presented multi-objective assembly line problem. The proposed algorithm considered Pareto concepts, used different neighbours of food sources for each employee bee and used crossover and mutation operation in its structure. Computational experiments are performed to solve standard assembly line benchmark problems taken from operations research (OR) library with the presented algorithm. The performance of proposed HPABC algorithm is compared with a famous multi-objective algorithm (SPEA 2) in literature. Computational results indicate that the presented HPABC algorithm outperforms SPEA 2 algorithm in most of the instances of the tested benchmark problems.
Disclosure statement
No potential conflict of interest was reported by the authors.