Performance up-gradation of Symbiotic Organisms Search by Backtracking Search Algorithm

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Swarm Intelligence

Background:

• Traditional Symbiotic Organisms Search (SOS) suffers from slow convergence and local optima stagnation.
• Backtracking Search Algorithm (BSA) is an alternative optimization method for complex problems.

Purpose of the Study:

• To address the limitations of SOS by proposing an enhanced ensemble algorithm, e-SOSBSA.
• To improve the intensification and diversification capabilities of optimization algorithms.

Main Methods:

• Incorporating BSA's self-adaptive mutation operator into SOS to escape local optima.
• Integrating BSA's adaptive crossover operator to explore previously visited search regions effectively.
• Ensemble of Symbiotic Organisms Search and Backtracking Search Algorithm (e-SOSBSA).

Main Results:

• e-SOSBSA demonstrated superior performance compared to component algorithms and state-of-the-art methods across various benchmark functions (CEC2014, CEC2015, CEC2017, CEC2020).
• Statistical, convergence, and diversity analyses confirmed the enhanced search capabilities of e-SOSBSA.
• The algorithm proved effective in solving seven real-world engineering optimization problems.

Conclusions:

• The proposed e-SOSBSA algorithm significantly overcomes the drawbacks of traditional SOS.
• e-SOSBSA offers a robust and effective approach for complex optimization tasks, including engineering applications.