An approximate solution method for general 0–1 programming problems through genetic algorithms with double string representation

In this paper, genetic algorithms using double string representation proposed by M. Sakawa et al. for 0–1 programming problems of the knapsack type are extended to be applied to more general 0–1 programming problems. Especially, for individuals represented by double strings, a new decoding algorithm including backtracking and individual modification is proposed to map each individual to a feasible solution. Furthermore, the efficiency and effectiveness of the proposed method will be investigated by comparing it with a branch-and-bound method concerning the accuracy of an approximate solution and the processing time through various numerical experiments.