Coarse-grained parallel genetic algorithm applied to a vector based land use allocation optimization problem: the case study of Tongzhou Newtown, Beijing, China

A Coarse-Grained Parallel Genetic Algorithm (CGPGA) is utilized to search for near-optimal solutions for land use allocation optimization problems in the context of multiple objectives and constraints. Plans are obtained based on the trade-off among three spatial objectives including ecological benefit, accessibility and compatibility. The Multi-objective Optimization of Land Use model integrates these objectives with the fitness function assessed by reference point method (goal programming). The CGPGA, as the first coupling in land use allocation optimization problems, is tested through the experiments with one processor, two processors and four processors to pursue near-optimal land use allocation scenarios and the comparison to these experiments based on Generic Genetic Algorithm (GGA), which clearly shows the robustness of the model we proposed as well as its better performance. Furthermore, the successful convergent (near-convergent) case study utilizing the CGPGA in Tongzhou Newtown, Beijing, China evinces the capability and potential of CGPGA in solving land use allocation optimization problems with better efficiency and effectiveness than GGA.