Modified firefly algorithm using randomized mechanisms

The firefly algorithm is a stochastic meta-heuristic algorithm that incorporates randomness into a search process. In essence, the randomness is useful when determining the next point in the search space and therefore has a crucial impact when exploring the new solution. Simultaneously, randomized mechanism plays an important role in balance the exploration and exploitation during the process. In this paper, an extensive comparison is made between 8 different probability distributions that can be used for randomizing the firefly algorithm's attractive mechanism, e.g., Uniform distribution, Gaussian distribution, Exponential distribution, Cauchy distribution, and so on. In our experiments, variously randomized firefly algorithms are developed and extensive experiments are conducted on 13-benchmark functions. The results of these experiments show that these randomized mechanisms can improve the convergence rate and the robustness of the firefly algorithm significantly.