Evaluation of a new modified Firefly algorithm

Abstract

Swarm based algorithms with features like easy implementation; self-organization and decentralization have led to emergent behavior successful in bridging the gap from idealistic situation to reality. Firefly Algorithm is one such new nature-inspired metaheuristic algorithms used for solving various optimization problems. It mimics the flashing behaviour of fireflies at night and was first proposed in 2008. In this work, performance of a new modified version of Firefly algorithm (MoFA) is compared with standard FA and two more versions of FA on various parameters. Results demonstrate that the new modified algorithm is superior to all the other three algorithms in all aspects.