Fitness Distance Balance Based Kepler Optimization Algorithm

Meta-Heuristic Search (MHS) algorithms are methods that continue to be popular and continue to be developed day by day in solving complex and high-dimensional global optimization problems. Kepler Optimization Algorithm (KOA) is an up-to-date MHS algorithm created by considering Kepler's laws for determining the position and velocity of planets. In the optimization process of the KOA, the position of each planet indicates a possible solution candidate, while the best solution is expressed as the sun. Obtaining the best solution using MHS algorithms in any optimization problem depends on avoiding the local solution traps in the search area. The equilibrium between exploration and exploitation is crucial for the diversity of solution candidates. In achieving this balance, the success of the search process depends on the reference position of the solution candidates determined by the selection method. Fitness Distance Balance (FDB) is a powerful selection method for determining the reference positions that guide the search process. Using the FDB, the most promising solution candidates for improving the population's search process are determined. In this study, a new FDBKOA is presented using the FDB method to increase the effectiveness of the KOA, avoid local solution traps, and develop the global optimum solution. Then, improved FDBKOA is compared with KOA in solving four challenging benchmark problems such as CEC2014, CEC2017, CEC2020, and CEC2022 for various problem types. When the results obtained are evaluated, it is seen that the FDB method increases the performance of KOA, and the FDBKOA algorithm gives more effective solutions than KOA in experimental studies.