Minimising material handling cost in cell formation with alternative processing routes by grouping genetic algorithm

Abstract

The alternative processing route is one of the important design factors for the cell formation problem (CFP) in cellular manufacturing systems (CMSs). Genetic algorithm (GA) is a popular method for solving the CFPs, because GA is capable of searching large regions of the solution's space while being less susceptible to getting trapped in local optima. However, the disadvantage of classical GAs is that the number of manufacturing cells should be known in advance. Knowing the actual number of manufacturing cells is relatively difficult before the CMS design is determined. Grouping genetic algorithm (GGA) is capable of solving CFP without predetermination of the number of cells, which is introduced by Falkenauer's GGA (1998). In order to adopt the GGA on CFP with alternative processing routes, we develop a new chromosome representation, a local optimisation algorithm for crossover operator and special mutation operators. These efforts ensure the efficiency of our method and are indicated in the numerical examples, and improved solutions are also obtained in the numerical examples.