Running city game optimizer: a game-based metaheuristic optimization algorithm for global optimization

As science and technology improve, more and more complex global optimization difficulties arise in real-life situations. Finding the most perfect approximation and optimal solution using conventional numerical methods is intractable. Metaheuristic optimization approaches may be effective in achieving powerful global optimal solutions for these complex global optimization situations. Therefore, this paper proposes a new game-based algorithm called the running city game optimizer (RCGO), which mimics the game participant's activity of playing the running city game. The RCGO is mathematically established by three newfangled search strategies: siege, defensive, and eliminated selection. The performance of the proposed RCGO algorithm in optimization is comprehensively evaluated on a set of 76 benchmark problems and 8 engineering optimization scenarios. Statistical and comparative results show that RCGO is more competitive with other state-of-the-art competing approaches in terms of solution quality and convergence efficiency, which stems from a proper balance between exploration and exploitation. Additionally, in the case of engineering optimization scenarios, the proposed RCGO is able to deliver superior fitting and occasionally competitive outcomes in optimization applications. Thus, the proposed RCGO is a viable optimization tool to easily and efficiently handle various optimization problems.