Salp Swarm Optimization: A Comprehensive Review of Recent Advances, Variants, Applications, and Future Research Directions

Abstract

One natural metaheuristic optimization technique is the Salp Swarm Algorithm (SSA), which takes its behavior from the swarming and feeding habits of salps in the ocean. With its straightforward structure, few parameters need, efficient exploration/exploitation balance, and adaptability to many optimization domains, SSA has garnered a lot of interest since its introduction. This paper provides a comprehensive review of SSA, highlighting its key advantages and applicability across a wide search space. The study includes its limitations, including sensitivity to problem types and reliance on the No Free Lunch theorem. This review analyzes different adaptations of SSA, such as binary versions, hybrid models, multi-objective extensions, and parameterless approaches, with the goal of enhancing performance and overcoming the limitations of the original algorithm. This paper analyzes the diverse applications of SSA in several domains, such as machine learning (including feature selection and neural network training), engineering optimization (covering scheduling, power systems, and renewable energy), image processing, localization, and additional practical areas. This study evaluates SSA through an analysis of its strengths, weaknesses, and potential areas for improvement. The study demonstrates that SSA is a promising and versatile optimization technique; however, it requires ongoing refinement to effectively address complex, dynamic, and multi-objective problems in future research.