Salp Swarm Algorithm Applications in Global Optimization

The Salp Swarm Algorithm (SSA) is a population-based metaheuristic inspired by the chain-forming behaviour of salp populations in the ocean. It models a leader–follower dynamic whereby the leader guides the search towards promising regions of the solution space while followers update their positions relative to upstream agents. This simple structure, combined with a small number of control parameters, yields a powerful balance between global exploration and local exploitation. Since its introduction, SSA has been applied to a wide range of global optimisation challenges, including continuous and constrained function optimisation, high-dimensional benchmark problems, engineering design, feature selection, neural network training and parameter estimation. Practical applications span civil and environmental engineering, machine learning, big data clustering and resource allocation, demonstrating its versatility and global significance in solving complex real-world problems.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent developments have focused on enhancing SSA performance on challenging global optimisation tasks. One approach integrates a Lévy flight mechanism with a sine–cosine operator to form an improved Salp Swarm Algorithm. This method employs long-range jumps to boost global exploration and adaptively switches between sine and cosine functions for precise local exploitation. Benchmark tests and neural network training demonstrate accelerated convergence and higher accuracy on high-dimensional problems. A second line of research introduces an opposition-based chaotic variant. By generating opposite candidate solutions and applying chaotic local search, this algorithm expands search diversity and avoids premature stagnation. Comparative studies on multimodal and unimodal test functions indicate superior convergence speed and solution quality over the original SSA and peer algorithms. More recently, the Salp Swarm Algorithm with a Local Escaping Operator (SSALEO) has been proposed for large-scale optimisation. Incorporating a population-diversity operator to escape local optima and a refined learning scheme enhances both exploration and exploitation. Evaluated on standard large-scale benchmarks, this variant outperforms classical SSA and several state-of-the-art optimisers, affirming its robustness for high-dimension problems.

Salp Swarm Algorithm Applications in Global Optimization publication trend

The graph below shows the total number of articles in salp swarm algorithm applications in global optimization across all publications each year (not limited to Nature Index journals).

Technical terms

Swarm intelligence: Computational paradigm modelling collective behaviour of decentralized agents to solve optimisation problems.

Exploration: Search process that investigates diverse areas of the solution space to avoid premature convergence.

Exploitation: Search process that intensively refines solutions near known good positions to enhance accuracy.

Lévy flight: Random walk characterised by heavy-tailed step lengths enabling occasional long jumps in the search space.

Opposition-based learning: Strategy generating opposite candidate solutions to improve convergence speed and population diversity.

Local escaping operator: Mechanism that perturbs solutions to help the algorithm escape local optima and enhance global search.

References

1. Improved Salp Swarm Algorithm Based on Levy Flight and Sine Cosine Operator. IEEE Access (2020).
2. An Opposition-Based Chaotic Salp Swarm Algorithm for Global Optimization. IEEE Access (2020).
3. Comparing SSALEO as a Scalable Large Scale Global Optimization Algorithm to High-Performance Algorithms for Real-World Constrained Optimization Benchmark. IEEE Access (2022).