The Firefly Optimization Algorithm: Convergence Analysis and Parameter Selection

Abstract

The bio-inspired optimization techniques have obtained great attention in recent years due to its robustness, simplicity and efficiency to solve complex optimization problems. The firefly Optimization (FA or FFA) algorithm is an optimization method with these features. The algorithm is inspired by the flashing behavior of fireflies. In the algorithm, randomly generated solutions will be considered as fireflies, and brightness is assigned depending on their performance on the objective function. The algorithm is analyzed on basis of performance and success rate using five standard benchmark functions by which guidelines of parameter selection are derived. The tradeoff between exploration and exploitation is illustrated and discussed.

Cuckoo Search in Test Case Generation and Conforming Optimality Using Firefly Algorithm

“…Then various tests are done which show that the newly proposed method performs better than many other algorithms. Arora et al [9] proved the recommendation of selecting the parameter on premises of rate of success and the performance by applying some functions. Analysis on convergence and the dynamic behavior of the firefly algorithm is done.…”

An Improved Optimized Web Page Classification using Firefly Algorithm with NB Classifier (WPCNB)

“…This is one of the first studies of gas is applied to the search domain of information. Chen and Kim (1995) proposed a hybrid system based on neural network genetic algorithm called MASCATO [13]. They used a GA to choose the best keywords that describe the documents selected by the user, and a neural network, weights keywords are determined.…”

Firefly Algorithm Hybridized with Flower Pollination Algorithm for Multimodal Functions

“…The light intensity I decreases with increase in distance r defined as I / 1 r 2 . Air acts as an absorbent medium that has the Firefly Algorithm Hybridized with Flower Pollination Algorithm …capability and absorbs the light in the medium that decreases the visibility of fireflies at some distance [8]. FA is based upon three rules that are:…”

“…These algorithms have received remarkable attention as they are known to be derivative free, robust, and can be applicable to different optimization problems [7]. These algorithms use the concept of randomization that has its efficiency moving away from local search to the global search [8].…”