Thesis (BS Applied Mathematics) -- University of the Philippines Mindanao, 2011

This study was done to solve uncapacitated facility location problems (FLP) using a new hybrid of two optimization algorithms. Particle Swarm Optimization-Invasive Weed Optimization (PSO--IWO) algorithm was applied to a real small-scaled data and a simulated large-scaled data. In this study, IWO was used as a local search heuristic t PSO which improved its searching ability towards the optimal solution. Twenty-four parameter sets were tested for the small-scaled data and twelve of these parameter sets were applied to the large-scaled data to verify results obtained from the small-scaled uncapacitated FLP. The solution obtained was further verified using binary integer programming. The optimal solution was to open facilities 3,4,6,7,8 and 10 with a total cost of 13.90518 million. It was observed that the maximum number of iterations had little or no effect on the algorithm since it was able to converge at an earlier time. The behavior of PSO-IWO observed in the small-scaled uncapacitated FLP was also evident in the large-scaled uncapacitated FLP. The optimal solution obtained incurred in a total cost of 263.15322 million. It was also observed that by not restricting the problem to open a limited number of facilities, a better solution can be obtained.