Decision support tool for a location problem based on a genetic algorithm

This paper considers the problem of locating the GDF (Gaz de France) bases of the French gas board in a geographical region in order to supervise and to maintain a set of stations and gas pipe lines with specific security and efficiency levels where some GDF base locations can be fixed a priori. The location problem reduces to a classical location allocation problem with additional side and security constraints. To solve the location problem we propose a genetic algorithm where solutions are evaluated based on their cost, their security and their efficiency levels. The genetic algorithm is embedded within a decision support tool to generate acceptable solutions to the decision maker. The decision support tool allows the decision maker to be involved in the decision process by setting the parameters of the location problem and the parameters of the genetic algorithm such as the initial population and the stopping criterion.