Application of Multi-objective Genetic Algorithms in Industrial Units Site Selection

The increasing growth of population, and hence, needing an optimal management of resources, inevitably makes land-use planning more important than before. Apparently, a suitable location for certain applications including industrial units may be involved in economic as well as social and ecological prosperities of the given region. Among the main problems on the industrial site selection, is the variety of conflicting economic and social factors, as well as the ecological ones, such that the existing functions and methods in GIS (buffer, overlay, etc.) can not meet the requirements of the multi-objectives plans. Hence the present research has been carried out to explore the suitable locations for large industrial units’ establishment. Having considered the multi-objective optimization methods, the authors applied Non-dominated Sorting Genetic Algorithm-II (NSGA-II) multi-objective genetic algorithm to support the decision making process as for industrial units site selection in Zanjan province. There fore, at the first step, objective functions and constrains have been defined and required data have been gathered from the key sources. Subsequently, the suitable lands have been determined and ranked, using NSGA-II method and GIS softwares as well as the specific software designed for the mentioned purpose. Results include the Optimum Location of Sites for large industries and the Pareto-front that shows the managers results of changes in the decision-making priorities. Also results of this algorithm have been tested in different conditions and showed the acceptable level of stability and repeatability of the algorithm.