Reinforcement learning-assisted evolutionary algorithm: A survey and research opportunities

Swarm and Evolutionary Computation, Год журнала: 2024, Номер 86, С. 101517 - 101517

Опубликована: Фев. 27, 2024

Язык: Английский

---

Machine Learning to Solve Vehicle Routing Problems: A Survey

IEEE Transactions on Intelligent Transportation Systems, Год журнала: 2024, Номер 25(6), С. 4754 - 4772

Опубликована: Янв. 2, 2024

This paper provides a systematic overview of machine learning methods applied to solve NP-hard Vehicle Routing Problems (VRPs). Recently, there has been great interest from both the and operations research communities in solving VRPs either through pure or by combining them with traditional handcrafted heuristics. We present taxonomy studies on paradigms, solution structures, underlying models, algorithms. Detailed results state-of-the-art are presented, demonstrating their competitiveness approaches. The survey highlights advantages learning-based models that aim exploit symmetry VRP solutions. outlines future directions incorporate solutions address challenges modern transportation systems.

Язык: Английский

---

Q-learning based multi-objective immune algorithm for fuzzy flexible job shop scheduling problem considering dynamic disruptions

Swarm and Evolutionary Computation, Год журнала: 2023, Номер 83, С. 101414 - 101414

Опубликована: Окт. 6, 2023

Язык: Английский

---

Collaboration and resource sharing in the multidepot time-dependent vehicle routing problem with time windows

Transportation Research Part E Logistics and Transportation Review, Год журнала: 2024, Номер 192, С. 103798 - 103798

Опубликована: Сен. 30, 2024

Язык: Английский

---

A recent review of solution approaches for green vehicle routing problem and its variants

Operations Research Perspectives, Год журнала: 2024, Номер 12, С. 100303 - 100303

Опубликована: Апрель 28, 2024

The green vehicle routing problem (GVRP) has been a prominent topic in the literature on logistics and transportation, leading to extensive research previous review studies covering various aspects. Operations seen development of exact approximation approaches for different extensions GVRP. This paper presents an up-to-date thorough GVRP spanning from 2016 2023, encompassing 458 papers. significant contribution lies updated solution algorithms applied both single-objective multi-objective Notably, 92.58% papers introduced mathematical model GVRP, with many researchers adopting mixed integer linear programming as preferred modeling approach. findings indicate that metaheuristics hybrid are most employed addressing Among approaches, combination metaheuristics-metaheuristics is particularly favored by researchers. Furthermore, large neighborhood search (LNS) its variants (especially adaptive search) emerges widely adopted algorithm These proposed within metaheuristic where A-/LNS often combined other algorithms. Conversely, predominant NSGA-II being frequently algorithm. Researchers utilize GAMS CPLEX optimization software solvers. MATLAB commonly language implementing

Язык: Английский

---

Q-Learning-Driven Butterfly Optimization Algorithm for Green Vehicle Routing Problem Considering Customer Preference

Biomimetics, Год журнала: 2025, Номер 10(1), С. 57 - 57

Опубликована: Янв. 15, 2025

This paper proposes a Q-learning-driven butterfly optimization algorithm (QLBOA) by integrating the Q-learning mechanism of reinforcement learning into (BOA). In order to improve overall ability algorithm, enhance accuracy, and prevent from falling local optimum, Gaussian mutation with dynamic variance was introduced, migration also used population diversity algorithm. Eighteen benchmark functions were compare proposed method five classical metaheuristic algorithms three BOA variable methods. The QLBOA solve green vehicle routing problem time windows considering customer preferences. influence decision makers’ subjective preferences weight factors on fuel consumption, carbon emissions, penalty cost, total cost are analyzed. Compared algorithms, experimental results show that has generally superior performance.

Язык: Английский

---

Hybrid Multi-Objective Artificial Bee Colony for Flexible Assembly Job Shop with Learning Effect

Mathematics, Год журнала: 2025, Номер 13(3), С. 472 - 472

Опубликована: Янв. 31, 2025

The flexible job shop scheduling problem is a typical and complex combinatorial optimization problem. In recent years, the assembly in problems has been widely studied. However, most of studies ignore learning effect workers, which may lead to higher costs than necessary. This paper considers with (FAJSPLE) proposes hybrid multi-objective artificial bee colony (HMABC) algorithm solve Firstly, mixed integer linear programming model developed where maximum completion time (makespan), total energy consumption cost are optimized simultaneously. Secondly, critical path-based mutation strategy was designed dynamically adjust level workers according characteristics path. Finally, local search capability enhanced by combining simulated annealing (SA), four operators different neighborhood structures designed. By comparative analysis on scales instances, proposed reduces 55.8 958.99 average over comparison algorithms for GD IGD metrics, respectively; C-metric, improves 0.036 algorithms.

Язык: Английский

---

A Q-learning-based multi-objective evolutionary algorithm for integrated green production and distribution scheduling problems

Engineering Applications of Artificial Intelligence, Год журнала: 2023, Номер 127, С. 107434 - 107434

Опубликована: Ноя. 11, 2023

Язык: Английский

---

Multi-objective reinforcement learning-based approach for pressurized water reactor optimization

Annals of Nuclear Energy, Год журнала: 2024, Номер 205, С. 110582 - 110582

Опубликована: Май 3, 2024

Язык: Английский

---

A Q-Learning Based Hybrid Meta-Heuristic for Integrated Scheduling of Disassembly and Reprocessing Processes Considering Product Structures and Stochasticity

Complex System Modeling and Simulation, Год журнала: 2024, Номер 4(2), С. 184 - 209

Опубликована: Июнь 1, 2024

Remanufacturing is regarded as a sustainable manufacturing paradigm of energy conservation and environment protection. To improve the efficiency remanufacturing process, this work investigates an integrated scheduling problem for disassembly reprocessing in where product structures uncertainty are taken into account. First, stochastic programming model developed to minimize maximum completion time (makespan). Second, Q-learning based hybrid meta-heuristic (Q-HMH) specially devised. In each iteration, method employed adaptively choose premium algorithm from four candidate ones, including genetic (GA), artificial bee colony (ABC), shuffled frog-leaping (SFLA), simulated annealing (SA) methods. At last, simulation experiments carried out by using sixteen instances with different scales, three state-of-the-art algorithms literature exact solver CPLEX chosen comparisons. By analyzing results average relative percentage deviation (RPD) metric, we find that Q-HMH outperforms its rivals 9.79%-26.76%. The comparisons verify excellent competitiveness solving concerned problems.

Язык: Английский

---