Optimization of Combat Resource Allocation Based on Restricted Tournament Selection Social Genetic Algorithm

Abstract To tackle the challenge of combat resource allocation problem (CRAP), especially under resource constraints, the dilemma between the efficiency of combat resource utilization and the efficiency of problem-solving. We propose a novel genetic algorithm that integrates a restricted tournament...

Full description

Saved in:
Bibliographic Details
Main Authors: Shandong Yuan, Yun Ren, Han Zhou, Yongjing Cheng, Kai Yan
Format: Article
Language:English
Published: Springer 2025-08-01
Series:International Journal of Computational Intelligence Systems
Subjects:
Online Access:https://doi.org/10.1007/s44196-025-00958-6
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract To tackle the challenge of combat resource allocation problem (CRAP), especially under resource constraints, the dilemma between the efficiency of combat resource utilization and the efficiency of problem-solving. We propose a novel genetic algorithm that integrates a restricted tournament selection strategy with sociological principles, named the restricted tournament selection social genetic algorithm (RTS2GA). Initially, we develop a comprehensive model for the allocation of combat resources that takes into account multiple constraints. Then, building upon the traditional Genetic Algorithm, we introduce a novel selection strategy known as restricted tournament selection to enhance the diversity of the algorithm’s population. In addition, we innovatively incorporate the concept of ‘group effect’ from sociology, adding a socialization operator to the algorithm to accelerate convergence and improve the quality of optimal solutions. Comprehensive evaluation confirms RTS2GA’s trade-off profile: though incurring added computational costs, it achieves competitive convergence speed (marginally behind PSO/MPSO; comparable to GA/DE/GA-APSO) while establishing definitive superiority in global optimization across all five benchmarks.
ISSN:1875-6883