Obtaining industrial experimental designs using a heuristic technique

There are several methods of obtaining orthogonal fractions of symmetrical and asymmetrical factorials, including Hadamard matrices, collapsing and replacing, and orthogonal arrays. The objective is to find the right permutations of levels of each factor in order to obtain uncorrelated main effects with a minimum number of runs. This leads to a combinatorial optimization problem. Simulated annealing is an optimization technique that has been successfully used to find a good solution for combinatorial optimization problems. By applying the computational technique of simulated annealing to orthogonal fractional factorial designs, we provide a method to obtain uncorrelated main effects for symmetrical and asymmetrical factorials.