Owing to the recent developments in the field of industrial automation, assembly lines have played an integral role in the economic uplift of the industrial units. Mixed Model Assembly Lines are the answer to a variety of scenarios which involve customized production following a particular ‘product mix’, i.e., several models of a product are jointly processed on a line, in an increased quantity, quality and productive environment. Hence, to determine the optimal operating schedule/sequence of the operations along with other impacting factors such as total utility work, setup cost, part consumption rates, etc., still remains a widely researched topic today. Moreover, sequencing problems are termed as NP-hard and a variety of sequencing heuristics have been applied in literature to solve them. The heuristic, Genetic Algorithm, was formulated based on binary encoding/decoding, two point cross over and uniform mutation, and applied in this paper to optimize two objectives; one, to minimize total utility work and two, to generate sequence of the models as per the first goal. A methodology was hence, developed to test and analyze the impact of factors such as number of stations, length of stations, conveyor speed, time of operations, number of primary models, and Minimum Part Set on the concerned objectives. An attempt was also made to model the entire process with IDEF0 modeling technique. Industry-oriented problems were then presented to test the algorithm in real world conditions. Finally, the results were critically examined and respective improvement measures were stated.
Genetic Algorithm for Concurrent Balancing of Mixed-Model Assembly Lines with Original Task Times of Models
