The multi-agent task allocation problem has attracted substantial research interest due to the increasing demand for effective solutions in large-scale and dynamic environments. Despite advancements, traditional algorithms often fall short in optimizing efficiency and adaptability within complex scenarios. To address this shortcoming, we propose a genetic algorithm-enhanced PPO (GAPPO) algorithm, specifically developed to enhance decision-making in challenging task allocation contexts. GAPPO employs a deep reinforcement learning framework, enabling each agent to independently evaluate its surroundings, manage energy resources, and adaptively adjust task allocations in response to evolving conditions. Through iterative refinement, GAPPO achieves balanced task distribution and minimizes energy consumption across diverse configurations. Comprehensive simulations demonstrate that GAPPO consistently outperforms traditional algorithms, resulting in reduced task completion time and heightened energy efficiency. Our findings underscore GAPPO’s potential as a robust solution for real-time multi-agent task allocation.

The multi-agent task allocation problem has attracted substantial research interest due to the increasing demand for effective solutions in large-scale and dynamic environments. Despite advancements, traditional algorithms often fall short in optimizing efficiency and adaptability within complex scenarios. To address this shortcoming, we propose a genetic algorithm-enhanced PPO (GAPPO) algorithm, specifically developed to enhance decision-making in challenging task allocation contexts. GAPPO employs a deep reinforcement learning framework, enabling each agent to independently evaluate its surroundings, manage energy resources, and adaptively adjust task allocations in response to evolving conditions. Through iterative refinement, GAPPO achieves balanced task distribution and minimizes energy consumption across diverse configurations. Comprehensive simulations demonstrate that GAPPO consistently outperforms traditional algorithms, resulting in reduced task completion time and heightened energy efficiency. Those findings underscore GAPPO’s potential as a robust solution for real-time multi-agent task allocation.

Multi-objective fuzzy Flexible Job shop Scheduling Problems (MofFJSPs) aim to optimize multiple fuzzy and conflicting objectives by finding the sequences of jobs and machines under realistic production constraints. However, as the objective functions of MofFJSPs exhibit fuzzy and conflicting properties, algorithms often face the challenge of being trapped in local optima and slow convergence. Therefore, to solve MofFJSPs effectively and efficiently, a Multi-Level Learning-aided Co-evolutionary Particle Swarm Optimization (MLL-CPSO) algorithm is proposed. The MLL-CPSO uses multiple populations to cooperatively solve multiple fuzzy objectives and hierarchically learns the evolutionary information, together with three novel designs. First, a Multi-Level Learning (MLL) strategy is proposed to learn the short-term personal evolutionary information, the long-term social information, and the co-evolutionary information to avoid local optima and approach the Pareto optima fast. Second, a Simulated Annealing-based Strengthening Diversity (SASD) strategy is designed to improve the diversity of populations and enhance the global search ability. Thirdly, a Co-evolutionary Information Update (CeIU) mechanism is developed to improve the quality of co-evolutionary information from different populations and drive populations to explore more Pareto optimal solutions. In experiments, MLL-CPSO is compared with 7 state-of-the-art algorithms on 3 typical benchmarks with 23 instances. The experimental results show that MLL-CPSO outperforms the compared algorithms in most test environments.