Abstract: Chimp Optimization Algorithm (ChOA) is one of the most efcient recent optimization algorithms, which proved its ability to deal with diferent problems in various do- mains. However, ChOA sufers from the weakness of the local search technique which leads to a loss of diversity, getting stuck in a local minimum, and procuring premature convergence. In response to these defects, this paper proposes an improved ChOA algorithm based on using Opposition-based learning (OBL) to enhance the choice of better solutions, written as OChOA. Then, utilizing Reinforcement Learning (RL) to improve the local research technique of OChOA, called RLOChOA. This way efectively avoids the algorithm falling into local optimum. The performance of the proposed RLOChOA algorithm is evaluated using the Friedman rank test on a set of CEC 2015 and CEC 2017 benchmark functions problems and a set of CEC 2011 real-world problems. Numerical results and statistical experiments show that RLOChOA provides better solution quality, convergence accuracy and stability compared with other state-of-the-art algorithms.

Key words: Chimp optimization algorithm , · Reinforcement learning , · Disruption operator , · Opposition-based learning , · CEC 2011 real-world problems , · CEC 2015 and CEC 2017 benchmark functions problems