Application of Genetic Algorithms in Mixed Model Assembly Line Balancing

Abstract

Mixed model assembly lines are a type of production line where a variety of product models withsimilar product characteristics are assembled. Line Balancing Problems are important for an efficient use ofmixed model assembly lines. This research introduces the use of artificial-intelligence based technique, so-calledgenetic algorithms (GAs), to solve mixed model assembly line balancing problems. Two important objectives ofassembly line balancing problems are considered simultaneously including minimizing number of workstationsand minimizing total idle time.

Experimental design are set up to test the significance of several parameters of GA including problemsizes, population sizes, crossover types, probability of cross-over, and probability of mutation. The results showthat the factors that significantly affect the performance of GAs are population size, crossover type and probabilityof mutation. As a result, it is necessary to define appropriate parameters while using GAs. However, the suitableparameters obtained from the research can be used as a guideline in practice. The performance comparisonbetween the proposed GAs and the known heuristic technique (COMSOAL) indicates that GAs performs significantlybetter than COMSOAL. From the research, it is found that GAs are powerful and efficient method that cansearch for a good solution within an acceptable time limit.

Keywords : Genetic Algorithms / Mixed Model Assembly Line BalancingAbstract1 Assistant