Assembly Line Balancing (ALB) is an attempt to assign tasks to various workstations along a line so that the precedence relations are satisfied and some performanceudmeasures are optimised. In this research, a few tasks that use similar resources will be assigned in the same workstation by ensuring that it does not violate the precedence constraint and that the total processing time in each workstation is approximately the same and does not exceed the cycle time. Assumption by previous researches that any assembly task can be performed in any workstation encourages the author to focus on the resource usage in ALB. Limited number of resources in the industry also becomes a vital influencer to consider this constraint in ALB. Apart from that, Elitist Non-Dominated Sorting Genetic Algorithm (NSGA-II) has not yet been implemented by previous researcher in the optimisation of Assembly Line Balancing Type-E (ALB-E) itself with resource constraints. The aim of this research is to establish a mathematical model for ALB-E with resource constraints (ALBE-RC). This research is proposed to be conducted in three main phases. After conducting literature review, the modelling phase will be performed. In the second phase of this research, an algorithm will be developed to optimise the problem. Later, the optimisation algorithm will be tested and verified using test problems from literature. The third phase of this research is, an industrial case study will be conducted for the purpose to validate the mathematical model and the optimisation algorithm. This research gap was identified when none of the previous research considered machine, tool, and worker constraint in ALB-E. In thisudresearch, a Genetic-based Algorithm was used as an optimisation approach. The Elitist Non-Dominated Sorting Genetic Algorithm (NSGA-II) has been proposed to optimiseudALBE-RC. The optimisation result indicated that the NSGA-II algorithm has better performance in finding non dominated solution due to small error ratio and small generational distance as compared to other algorithms like Multi-Objective Genetic Algorithm (MOGA) and Hybrid Genetic Algorithm (HGA). The results indicate that NSGA-II has the ability to explore the search space and has better accuracy of solution towards Pareto-optimal front. The validation phase from the industrial case study concluded that the proposed methodology and algorithm can be implemented in industries. The cycle time of existing layout had been extensively decreased from 16.1 seconds to 13.1 seconds after the optimisation. The number of workstations was decreased after the optimisation from 17 workstations to nine (9) workstations. Meanwhile, the number of resources used were reduced from 43 resources to 40 resources. Apart from that, the percentage of line efficiency improved from 33.8% to 78.4%. These results indicated that the developed methodology and the proposedudalgorithm can reduce the utilisation of resources, workstations and cycle time. In fact, the aforementioned approach also can increase the efficiency of assembly process as well as enhance the industrial productivity.
展开▼
