A DQN-Driven Multi-Objective Strike Sequence Optimization: An Enhanced Hyper-Heuristic Approach

This study addresses the optimization sequencing of critical nodes in enemy target systems during large-scale joint operations. We establish a multi-objective optimization model that balances strike effectiveness and resource consumption for target prioritization. Algorithmically, we propose an enhanced Non-dominated Sorting Genetic Algorithm II (NSGA-II), which integrates a Deep Q-Network (DQN) mechanism to strengthen local search capabilities and employs a multi-dimensional evaluation metric system to improve solution diversity. Extensive simulations demonstrate the algorithm's superior solution quality compared to conventional methods, offering a novel paradigm for target prioritization decision-making in large-scale combat operations.