Possibilities of Ga in Optimization of Manufacturing Cell Formation
چکیده
Optimization of manufacturing cell formation includes a whole virtual cell formation procedure and has NP-hard complexity. In the last decades, evolutionary algorithms have been widely applied to real-world optimization including to manufacturing cell formation problems. Genetic algorithms (GAs) are possibly the most widespread variant of evolutionary algorithms and are now frequently used for a number of optimization problems in operation management. In this study it is discussed about theoretical and practical aspects of GA to bridge the gap between hypothetical analysis and real time implementation of GA in a given domain. Keyword: cell formation, genetic algorithm, cellular manufacturing. INTRODUCTION In manufacturing firms, characterized by a high variety and low to medium volume product, mix the conventional layout approach is process oriented that is very flexible and insures high equipment utilization rates. An alternative to this traditional layout is the use of Cellular Manufacturing Systems (CMS) that combines the advantages of both product and process oriented systems. Cellular manufacturing (CM) is an application of the group technology (GT) philosophy to designing manufacturing systems. GT approach refers to grouping of parts to be manufactured according to similarities derived using various characteristics and processing them on their requisite machines placed close together in a group called a cell. In industry, it is not easy to make all parts and machines into stand alone cells, and therefore a separate shop to look after such kind of special operations becomes mandatory. There are firms having some operations outside the cells for different products and spare parts whose changing demand patterns tend to disturb the cell routine works. It has also been pointed out that as the range of parts becomes wider, with parts in different stages of their lifecycles, it is preferable for the manufacturing system to be configured in a hybrid manner [1]. In addition, if there is only one machine or facility of a certain type (such as painting, furnace equipment), used for processing a wide range of parts, it would be advantageous to retain such special machine types in the functional layout. The processing times of these machines are often significantly different, preventing them from working in harmony with other parts in the cells. The cost of duplicating these machines and placing them in multiple cells is often high. As a result, these special machine types (which may also be used by the stabledemand parts processed in the cells) may have to be located in the functional layout. There are number of objectives need to be solved in a real time environment. For example, grouping efficiency of the cells, inter-cell movement, logistics handling and so on. There have been numerous methods available in the literature including modern ones like Simulated Annealing (SA), Genetic Algorithms (GA), Tabu Search (TS), Greedy approaches, Variable-Depth search, Hill climbing procedures, and Ant Colony Optimization (ACO), which can be used to solve such objective functions theoretically. While going for real implementation in shop floor, how many of them do effectively work? In this paper a potential of genetic algorithms for manufacturing cell formation will be considered. Genetic Algorithm is a computerized search and optimization algorithm based on the mechanics of natural genetics and natural selection. GA is a search technique for global optimization in a search space. As the name suggests, they employ the concepts of natural selection and genetics using past information for directing the search with expected improved performance to achieve fairly consistent and reliable results. The traditional methods of optimization do not work sufficiently over a broad spectrum of problem domain. GA attempts to mimic the biological evolution process for discovering good solutions. They are based on a direct analogy to Darwinian natural selection and mutations in biological reproduction and belong to a category of heuristics known as randomized heuristics that employ randomized choice operators in their search strategy and do not depend on complete a priori knowledge of the features of domain. These operators have been conceived through abstractions of natural genetic mechanisms such as crossover and mutation and have been cast into algorithmic forms. Holland [2] envisaged the concept of these algorithms in the mid-sixties and it has been applied in diverse areas such as music generation, genetic synthesis, fault diagnosis, strategy planning and also to address business problems such as travelling salesman problem, production planning and scheduling problem, facility location problem, transportation problems, telecommunications and network problems, engineering design problems and image processing and cell design problems. GA is different from traditional optimization and search