Biogeography-Based Optimization with LéVY-Flight Exploration for Combinatorial Optimization

摘要

Biogeography Based Optimization (BBO) algorithm is one of the many evolutionary algorithms. BBO uses the concept of biogeography to find the optimal solution. It mainly consists of two operators, namely Migration and Mutation. In this paper, we have proposed a novel variant of BBO (LBBO, using ‘Lévy Flights’) by modifying its migration and mutation operator to improve their exploration capability. Migration operator is modified by considering the average value of best islands to update migrating islands while mutation operator is modified by applying lévy flight distribution for exploration. The proposed algorithm has been compared with other evolutionary algorithms like GA, PSO and the original BBO on ten benchmark functions. Simulation results show that the LBBO outperforms. We have also tested the performance of proposed LBBO on binary knapsack problem.