Solving Bi-level Problems Using Modified Particle Swarm Optimization Algorithm
Nancy El-Hefnawy · 2014
The bi-level programming problem (BLPP) is considered to be NP-hard problem. This paper suggests a modified particle swarm optimizer (MPSO) for solving fuzzy bi-level single and multi-objective problems. In this approach the BLPP handles as fuzzy multi-objective problem. Most traditional algorithms designed for specific versions or based on specific assumptions in the BLPP, thus it is difficult to improve the diversity and expand the search space of the particle. For such problems, new strategy for the adaptive inertia weight in PSO is proposed to control the domain of the particle oscillation according to fitness function, and to eliminate the need for velocity clamping. In addition, fuzzy utility membership function introduced to extract the best compromise solution based on fuzzy set theory. Finally, simulation results will be presented for six test problems and compare them with other algorithms to illustrate the efficiency of the proposed algorithm.