A NEW CROSSOVER TECHNIQUE IN GENETIC ALGORITHMS Mehrdad Kazerooni 1 , Afshin Kazerooni 2 , 1 Assistant Professor, Mechanical Engineering Dept., K.N.Toosi University of Technology, West-Mirdamad St., Tehran, Iran, Tel:0098-21-7333549, FAX: 0098-21-7334338 kazerooni@kntu.ac.ir 2 Assistant Professor, Mechanical Engineering Dept, Rajaei University, Tehran, Iran, Tel:0098-21-7333549, FAX: 0098-21-7334338 afshin@srttu.edu Abstract: Genetic algorithms have been used for many years to solve optimization problems. They have been employed for many engineering application such as computer aided process planning, scheduling, plant layout, cell formation, prediction, supply chain management and many others. Any genetic algorithm at least has four steps in a complete cycle. The selection step plays the most important role in any genetic algorithm. It consists of two sub steps, crossover and mutation. This paper describes the development of a new crossover technique called Advanced Edge Recombination (AER) to increase the efficiency of genetic algorithms for combinatorial problems including traveling salesman problem, cell formation and cellular layout problem. The results obtained by this new technique have been compared with other existing techniques to prove its efficiencies over them. Keywords: Genetic algorithms, crossover, TSP, plant layout, cellular manufacturing GENETIC ALGORITHMS Genetic methods seek to mirror the evolutionary processes observed in the natural world. At each iteration, a small sub-set of feasible solutions, called a generation, is manipulated in ways that produce a new sub-set of solutions that one hopes will be better than those of the previous generation. The manipulations include selection, reproduction (or crossover) and mutation. In selection, the best solutions of a generation are retained for further use and the weaker solutions are eliminated. In reproduction or crossover, two chosen solutions (parents) are combined so as to produce a new feasible solution (child) that retains the strengths of the parents. In mutation, random changes are made to a solution in order to increase variety and prevent premature convergence of the procedure. In some implementations, including the one presented here, random mutation is replaced by a local improvement method used to improve convergence rates. A solution is usually called a chromosome and the elements of a chromosome are usually called genes. These methods have been well documented in the literature, for example by Holland (1975), Goldberg (1989).They retained the attention of researchers because they permit the solution a wide variety of PDF created with pdfFactory Pro trial version www.pdffactory.com