Multi-Objective Optimization, Approach for Land Use Allocation Using NSGA-II

Land use pattern, due to increasing human activities and interference on land in order to meet different needs, is constantly changing. Most of land use optimization as a complex problem in the real world inherently involves multiple conflicting objectives. In this research, the capability of non dominated sorting genetic algorithm II (NSGA-II) with implying two type crossover operators, namely as two dimensional cross over (TDC) and uniform cross over (UC) in order to multi-objective optimization of land-use allocation has been evaluated in the Taleghan watershed. The multi-objective problem described in this paper consists of four objective functions: maximizing economic return, maximizing suitability, maximizing compactness-consistency, maximizing runoff and minimizing erosion. The model was implemented in two ways, with guidance and without guidance in generating the initial population. Results show that the crossover operator performance is completely under the influence of initial population. The comparison of the examined crossover operators shows that there is a significant advantage of using TDC operator in generating land use allocation patterns with respect to the given objective functions. The results showed that after the optimization in the study area, erosion rates declined while profits increased