Complicating features in industrial grouping problems

Due to a number of complicating features, industrial grouping problems are generally NP-hard and computationally difficult to comprehend and model. Based on the recent case studies, this chapter identified characteristic complicating features that pose challenges to decision makers concerned with grouping problems. These features were classified into model abstraction, presence of multiple constraints, fuzzy management goals, and computational complexity. Results of an in-depth taxonomic study of 18 case studies in the literature revealed a number of complicating features within the four categories. Among the methods that have been applied in these case studies, genetic algorithm is the most widely used. This indicated the great potential of the algorithm to solve a wide range of grouping problems. Realizing the inadequacies of solution methods applied, the study suggested the use of flexible, fuzzy multi-criteria grouping algorithms that hybridize fuzzy theory, fuzzy logic, grouping genetic algorithms, and intelligence. It is hoped that advances and applications of grouping genetic algorithm based on these techniques will yield remarkable progress in developing decision support tools for industrial grouping problems.