Damage Maximization for Combat Network with Limited Costs

TL;DR

This paper introduces a realistic model for damage maximization in combat networks, incorporating topology-based costs and effects, and proposes an improved genetic algorithm to find optimal attack strategies, aiding military decision-making.

Contribution

It presents a novel, practical damage maximization model for combat networks and an improved genetic algorithm tailored for this problem.

Findings

The proposed model effectively captures realistic damage costs and effects.

The improved genetic algorithm outperforms existing methods in feasibility and effectiveness.

Analysis of attack patterns and algorithm convergence enhances understanding of combat network vulnerabilities.

Abstract

Maximizing the damage by attacking specific nodes of the combat network can efficiently disrupt enemies' defense capability, protect our critical units, and enhance the resistance to the destruction of system-of-system~(SOS). However, the modeling of the combat network damage is not practical enough. In this paper, we report a more realistic model to study the combat network damage maximization problems. By analyzing realistic situations, the cost of damage is redefined based on the network topology and the functional characteristics of nodes. The damage effect is also updated according to the combat network topology and operational capability. Hence, a cost-limited damage maximization model for the combat network is constructed. In addition, to obtain optimal solutions, an improved genetic algorithm~(IPGA) based on prior information is proposed. As a result, our method has a significant advantage in the feasibility and effectiveness compared with other algorithms in experiments. The attack pattern of the combat network and the convergence and complexity of the proposed algorithm are further explored. The improved model and algorithm, as well as the mined attack patterns, can provide support for military decisions.