Comparative analysis and optimization of Mixed Model assembly line using Genetic Algorithm. (2018)

• Record Type: Journal Article
• Authors: Sahu, Abhishek; Pradhan, S.K.
• Is Part Of: Materials today. Volume 5:Number 11(2018)Part 3
• Page Start: 25075
• Page End: 25084
• Publication Date: 2018
• Subjects: Line balancing -- workstation -- genetic algorithm -- cyclic crossover -- ideal time -- bottleneck; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2018.10.308
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: Line balancing aims at arrangement of different facilities like workstation, machines, tool, and operator in an adequate manner in order to obtain most efficient balancing of assembly line, its capabilities, flow of product and services. Various methods are proposed in the literature to optimize these parameters hence this research effort starts with the review of the various methods developed and to compare main features of few of them for their relative assets and limitation to allow choose the most suitable method for optimization of line balancing. The researchers used the average times of the tasks of the models as the timings of the tasks of the combined assembly line model while designing the assembly line which is an approximate method of dealing with the data. Hence, in this paper, the design of the assembly line of the models is done simultaneously for all the models using their original task times. Such an attempt gives a realistic design of the mixed-model assembly line using genetic algorithm for the balancing of this mixed model assembly line problem. A new methodology of balancing the assembly line has been adopted and the output is compared with conventional methods of line balancing. The optimization of assembly line using cyclic multipoint crossover method of genetic algorithm has been performed to obtained real offspring and the best chromosome is selected to find out the efficiency of assembly line. The superiority of this approach is demonstrated using a numerical example. The multi-objective genetic algorithm with cyclic crossover and with individual task times of the models is a unique contribution. Although the RPW method also gives better results other than any conventional methods of line balancing however when the population size (assembly lines) of the system increase and number of workstation are more than genetic algorithm method is suitable and gives better result and this research paper presents the comparative analysis of mixed model assembly line balancing methods and tries to find out the best possible solution in order to minimize no. of workstation, ideal time, bottleneck and increase the efficiency of plant