Solving the distribution network routing problem with artificial immune systems

Successful planning of electrical distribution networks is a complex problem. When solving that problem, it is commonly translated into combinatorial optimization problems, like single and multiple depot vehicle routing problems (MDVRP). Such optimization problems are NP-hard, hence exact solving is practically impossible. Evolutionary algorithms have been successful in solving those problems. Although they are very efficient, expected progress is related to obtaining better stability and lesser dependency on parameters. In this paper, an improvement of genetic algorithm for solving the MDVRP is shown, inspired by artificial immune systems' techniques. Based upon the analogy of MDVRP and spatially closed distribution network planning problem, some practical examples have been used to investigate the performance of the proposed algorithm.