Age-Structured Biogeography-based Optimization

Abstract

Biogeography-based optimization (BBO) is a nature-inspired metaheuristic, motivated by the migration of species between habitats. It has been shown to provide good performance on several optimization tasks. However, BBO does not capture the effects of age structures of individuals. In the past, the concept of age has been incorporated by other nature- inspired algorithms, leading to a general improvement in their optimization capability. To reap similar benefits, a novel variant of BBO called age-structured BBO (ASBBO) is proposed, that introduces the effects of the age of individuals on the migration process. It is based on the observation that, in nature, young and old individuals are less suitable for migration in comparison to those of the middle-age. Individuals with birth rates and survival rates are introduced to capture age-related dynamics. Age structures allow information to be shared based not only on the current goodness of a solution, but also on its history, resulting in a more informed decision, and subsequently, an improved search capability. Computational experiments conducted on standard benchmark functions validate the superiority of ASBBO over BBO and three other have known nature-inspired optimization algorithms, namely, ant colony optimization (ACO), genetic algorithm (GA) and particle swarm optimization (PSO).