A GA-based search method for the tolerance allocation problem

This paper considers nonlinearly constrained tolerance allocation problems in which both tolerance and process selection are to be selected simultaneously so as to minimize the manufacturing cost. The tolerance allocation problem has been studied in the literature for decades, usually using mathematical programming or heuristic optimization approaches. The difficulties encountered for both methodologies are the number of constraints and the difficulty of satisfying the constraints. A penalty-guided genetic algorithm is presented for solving such mixed-integer tolerance allocation problems. It can efficiently and effectively search over promising feasible and infeasible regions to find the feasible optimal or near optimal solution. Genetic results are compared with the results obtained from 12 problems from the literature that dominate the previously mentioned solution techniques. Numerical examples indicate that the genetic algorithms perform well for the tolerance allocation problem considered in this paper. In particular, as reported, solutions obtained by genetic algorithms are as well as or better than the previously best-known solutions.