Many real-world applications for complex industrial engineering or design problems could be modelled as optimisation problems.
These problems are often characterised by complex features such as multi-modality, dynamics, discontinuity, and nonlinearity. Learning-driven methods have been identified as practical approaches for agents to learn the solution space by taking actions and interacting with the environment, thereby continuously updating their strategies. Reinforcement Learning (RL) and metaheuristic algorithms are emerging approaches, utilising advanced computation power to address such challenges. These approaches have been actively investigated and applied, particularly to scheduling and logistics operations.
Over the past several years, a substantial amount of research in evolutionary algorithm (EA) improvement has focused on integrating RL into the EA framework, referred to as reinforcement learning-assisted evolutionary algorithm (RL-EA). The RL-EA effectively leverages the acquired search information to optimise solutions collaboratively, demonstrating its success across various problem domains. Despite the successful application of RL-EA in many areas, the theoretical analysis of algorithms, benchmarks, training methods, and strategy design is still an open field of research. It still faces the challenge of high computational costs and sparse rewards. Therefore, there is a need to explore novel methods to enhance algorithm performance.
Learning-driven optimisation algorithms could be utilised to solve more general optimisation problems, especially for issues that are very difficult to solve with traditional hill-climbing algorithms. For scheduling problems, massive data are collected and used to optimize route selection, taxi dispatching, dynamic transit bus scheduling, and other mobility services to improve the efficiency of the operations. For logistics problems, material movements within and between supply chain entities, including warehouses, factories, distribution centres, and retail shops, are improved and optimized with advanced data-oriented techniques.
Due to the complexity of real-world applications, no one panacea could solve all troubles in real-world cases. Learning-driven and meta-heuristics methods are approaches that should be adapted to handle practical real-world scheduling and logistics applications.
Recently, the interaction between evolutionary computation algorithms and reinforcement learning has received considerable attention from the research community and the industry. The main aim of this Research Topic is to report the recent progress in integration methods of reinforcement learning-driven optimization algorithms for scheduling and logistics. This Research Topic will publish original research papers on theoretical analysis of algorithms, benchmarks, and training methods specific to this new class of techniques, parameter tuning, and real-world applications in advancing scheduling and logistics. Submissions involving real-world case studies are encouraged in the following topics (but not limited to):
• Reinforcement Learning,
• Deep Reinforcement Learning
• Bio-inspired algorithms,
• Nature-inspired Computing
• Computational Intelligence,
• Evolutionary Algorithms
• Metaheuristic Algorithms,
• Swarm Intelligence
• Agent-based Simulation,
• Multi-Agent Systems
• Intelligent Scheduling Systems,
• Decision Support Systems
• Intelligent Logistics Systems,
• Reverse Logistics Systems
• Multi-model Transport,
• Scheduling & Logistics
• Supply Chain (SC) Network,
• SC Management with Sustain. Devel. Goals
• Vehicle Routing Problems,
• Underground Logistics Systems,
• International Logistics Management
Keywords:
tbc
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
Many real-world applications for complex industrial engineering or design problems could be modelled as optimisation problems.
These problems are often characterised by complex features such as multi-modality, dynamics, discontinuity, and nonlinearity. Learning-driven methods have been identified as practical approaches for agents to learn the solution space by taking actions and interacting with the environment, thereby continuously updating their strategies. Reinforcement Learning (RL) and metaheuristic algorithms are emerging approaches, utilising advanced computation power to address such challenges. These approaches have been actively investigated and applied, particularly to scheduling and logistics operations.
Over the past several years, a substantial amount of research in evolutionary algorithm (EA) improvement has focused on integrating RL into the EA framework, referred to as reinforcement learning-assisted evolutionary algorithm (RL-EA). The RL-EA effectively leverages the acquired search information to optimise solutions collaboratively, demonstrating its success across various problem domains. Despite the successful application of RL-EA in many areas, the theoretical analysis of algorithms, benchmarks, training methods, and strategy design is still an open field of research. It still faces the challenge of high computational costs and sparse rewards. Therefore, there is a need to explore novel methods to enhance algorithm performance.
Learning-driven optimisation algorithms could be utilised to solve more general optimisation problems, especially for issues that are very difficult to solve with traditional hill-climbing algorithms. For scheduling problems, massive data are collected and used to optimize route selection, taxi dispatching, dynamic transit bus scheduling, and other mobility services to improve the efficiency of the operations. For logistics problems, material movements within and between supply chain entities, including warehouses, factories, distribution centres, and retail shops, are improved and optimized with advanced data-oriented techniques.
Due to the complexity of real-world applications, no one panacea could solve all troubles in real-world cases. Learning-driven and meta-heuristics methods are approaches that should be adapted to handle practical real-world scheduling and logistics applications.
Recently, the interaction between evolutionary computation algorithms and reinforcement learning has received considerable attention from the research community and the industry. The main aim of this Research Topic is to report the recent progress in integration methods of reinforcement learning-driven optimization algorithms for scheduling and logistics. This Research Topic will publish original research papers on theoretical analysis of algorithms, benchmarks, and training methods specific to this new class of techniques, parameter tuning, and real-world applications in advancing scheduling and logistics. Submissions involving real-world case studies are encouraged in the following topics (but not limited to):
• Reinforcement Learning,
• Deep Reinforcement Learning
• Bio-inspired algorithms,
• Nature-inspired Computing
• Computational Intelligence,
• Evolutionary Algorithms
• Metaheuristic Algorithms,
• Swarm Intelligence
• Agent-based Simulation,
• Multi-Agent Systems
• Intelligent Scheduling Systems,
• Decision Support Systems
• Intelligent Logistics Systems,
• Reverse Logistics Systems
• Multi-model Transport,
• Scheduling & Logistics
• Supply Chain (SC) Network,
• SC Management with Sustain. Devel. Goals
• Vehicle Routing Problems,
• Underground Logistics Systems,
• International Logistics Management
Keywords:
tbc
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.