Nature-Inspired-Metaheuristic-Search-Algorithms-for-Optimizing-Benchmark-Problems-Inclined-Planes-S

In the literature, different types of inclined planes system optimization (IPO) algorithms have been proposed and evaluated in various applications. Due to the large number of variants and applications, this work provides an overview of IPO’s state-of-the-art in terms of variants presented, applications, statistical evaluation, and analysis. In addition, the performance of IPO variants are evaluated and compared. The results are benchmarked against other algorithms. Final evaluation based on statistical analysis and a new and effective ranking methodology indicates the optimal performance and relative success of all IPO variants and their performance in comparison with other recent diverse metaheuristic search competitors, including reinforcement learning, evolution-based, swarm-based, physics-based, and human-based. The performance of IPO variants shown that the use of bio-operators to improve the standard version is more successful than other applied approaches. So that, the successful performance of SIPO + M with a minimum overall ranking of 0.73 has been ahead of all versions, and the complexity of IPO equations has also been led to a high time loss and achieving a maximum overall ranking of 2.07. Among other algorithms, it shown that versions without control parameters perform exploration and exploitation processes intelligently and more successful. For example, POA-I, POA-II, SLOA, OPA, and CMBO are among the methods that achieved the best performance, with minimum overall ranking values of 0.363, 0.384, 0.387, 0.424, and 0.933, respectively.

https://rdcu.be/cUAbj

https://doi.org/10.1007/s11831-022-09800-0