A combined genetic adaptive search (GeneAS) for engineering design

In this paper, a flexible yet efficient algorithm for solving engineering design optimization problems is presented. The algorithm is developed based on both binary-coded and realcoded genetic algorithms (GAs). Since both GAs are used, the variables involving discrete, continuous, and zero-one variables are handled quite efficiently. The algorithm restricts its search only to the permissible values of the variables, thereby reducing the search effort in converging to the optimum solution. The efficiency and ease of application of the proposed method is demonstrated by solving three different mechanical component design problems borrowed from the optimization literature. The proposed technique is compared with binarycoded genetic algorithms, Augmented Lagrange multiplier method, Branch and Bound method and Hooke and Jeeves pattern search method. In all cases, the solutions obtained using the proposed technique are superior than those obtained with other methods. These results are encouraging and suggest the use of the proposed technique to other engineering design problems.