-

In this paper, cellular manufacturing (CM) problems in dynamic and stochastic states have been solved by using a genetic algorithm. In fact, cellular manufacturing system is an application of group technology in a production system. The aim of such system is to cluster parts and machines into the related families in such a way that operational similarity in different production and planning aspects is considered. In most previous researches, cellular manufacturing problems in static production or deterministic demand have been considered continuously. However in a real-world situation, dynamic production and demand for products are non-deterministic. So in order to adapt a CM model with real situation, a great number of variables and restrictions are required. Thus, solution to the above model requires more computational
time, memory, and high processing power of
computer with traditional optimization methods. Consequently nowadays, modern heuristic methods such as genetic algorithms has been applied in a group of stochastic search techniques in order to solve such a NPcomplete problem. In this paper, first a nonlinear integer model is developed and then it is solved by genetic algorithms. Finally, obtained results are compared with optimal solutions to show the validity of the proposed algorithm.