Article Preview
Top1. Introduction
In the past few decades several nature inspired algorithms (NIA) have emerged for solving global optimization problems. NIA algorithms may be classified as the ones based on natural evolution, commonly known as Evolutionary Algorithms (EA) and the ones that are based on behavioral pattern displayed by various species, particularly the ones that live in groups (or swarms). Evolutionary Algorithms (EAs) are optimization techniques based on the concept of a population of individuals that evolve and improve their fitness through probabilistic operators like recombination and mutation. These individuals are evaluated and those that perform better are selected to compose the population in the next generation. After several generations these individuals improve their fitness as they explore the solution space for optimal value. Some popular EA are Genetic Algorithms (GA) (Goldberg, 1989; Holland, 1975), Evolutionary Strategies (ES) (Rechenberg, 1994), Evolutionary Programming (EP) (Back & Schwefel, 1996; Back, Hammel, & Schwefel, 1997; Fogel, 1995; Fogel, Owens, & Walsh, 1996), and Differential Evolution (DE) (Storn & Price, 1995, 1997).
The other class of NIA is based on the intelligent behavior exhibited by various species like fish birds, bees, ants, termites etc. The algorithms belonging to this class are also known as swarm algorithms and are modeled on the information sharing mechanism of the species. Some popular algorithms belonging to this class include Particle Swarm Optimization (PSO) (Kennedy & Eberhart, 1995), Ant Colony Optimization (ACO) (Dorigo & Gambardella, 1997), Artificial Bee Colony (ABC) algorithm (Karaboga & Basturk, 2007, 2008; Karaboga, 2005; Basturk & Karaboga, 2006), Bacteria Foraging Algorithm (BFO) (Passino, 2002), and Biogeography Based Algorithm (BBO) (Simon, 2008). A survey of algorithms based on bee swarm algorithms is given by Karaboga and Akay in (2009).
In the present study, the focus is on ABC, which is one of the recently proposed NIA. ABC follows the analogy of the socio-cooperative behavior demonstrated by honey bees in their search for nectar. A brief overview of the working of ABC algorithm is given in section 3.
ABC has been successfully applied for solving a variety of real life and benchmark problems. Its comparison with the contemporary algorithms has shown its competence in dealing with different types of problems (Singh, 2009; Karaboga, 2009; Ponton & Klemes, 1993; Rao, Narasimham, & Ramalingaraju, 2008; Karaboga, Akay, & Ozturk, 2007; Pawar, Rao, & Davim, 2008; Karaboga & Basturk, 2007; Pan, Tasgetiren, Suganthan, & Chua, 2011).
However, like most of the NIA in their basic form, ABC is not completely flawless. As pointed out by Zhu and Kwong (2010), ABC has a structure which supports exploration (diversification) more in comparison to exploitation (intensification). These two antagonist factors should be balanced judiciously for the success of an optimization algorithm. In the present study an attempt is made to include the Differential Evolution (DE) operators in the structure of basic ABC. The rationale is to improve the exploitation capabilities of ABC algorithm. In the proposed algorithm named as Differential Evolution Operators Embedded Artificial Bee Colony (DE-ABC), ABC is used to explore the solution space in search of positions of potential food sources while the DE operators are used to exploit those sources with the hope of finding a better location (or position) of food sources.