A NOVEL HYBRID BAT CROW SEARCH ALGORITHM FOR SOLVING OPTIMIZATION PROBLEMS

Özet

Meta heuristic algorithms like Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and other
algorithms are great and famous techniques used to solve many hard and complex optimization
problems. This paper presents a new hybrid algorithm named Hybrid Bat Crow Search Algorithm
(HBCSA). To achieve this algorithm, two algorithms were considered. The algorithms are Crow
Search Algorithm (CSA) and Bat Algorithm (BA). The advantageous points of the two algorithms
were taken into consideration and used to design an effective hybrid algorithm that can give
significantly high performance in many benchmark functions. In addition, quantum behaved PSO
equation is used in this hybrid algorithm. This leaded to better results when testing the algorithm
against Benchmark problems. The combination of concept and functionality of Bat and Crow
algorithms enable the suggested hybrid algorithm of making an appropriate trade-off between
exploration and exploitation capabilities of the new algorithm.
For the purpose of evaluating the performance of the new Hybrid Bat Crow Search Algorithm
(HBCSA), some well known Benchmark functions were utilized. In the new algorithm every member
in the swarm will have behave like a crow in the sense of observing other members in the swarm to
see where they hide their foods. In the same time, as in bats, every member will use echo system while
searching its own solution. Echo system is integrated with PSO equations. Each member has an
awareness parameter as in CSA. According to awareness parameter a member can know whether if
another member is following it or no. These are the basic lines of the new HBCSA. The results
indicated that the proposed HBCSA can produce very competitive solution when compared to other
famous and state of the art meta-heuristic algorithms.