A Reinforcement Learning-Driven Improved MOEA/D Algorithm for Configuration Optimization of Four-Way Shuttle Warehouse Systems
Year: 2026
Type: dataset
Abstract: Dataset Description:This dataset contains the experimental data, simulation results, and configuration parameters supporting the research on the Four-Way Shuttle Warehouse System (FWSWS).Specifically, this dataset includes:System Parameters: Basic configuration parameters of the warehouse system, including spatial layout limits, equipment resources, kinematic parameters, and cost coefficients used in the models.Algorithm Performance Metrics: The statistical results of Hypervolume (HV) and Inverted Generational Distance (IGD) values. These metrics compare the proposed RL-IMOEA/D algorithm against other standard optimization algorithms (such as MOEA/D, NSGA-II, NSGA-III, SPEA2, and MOPSO) across multiple independent runs.Pareto Optimal Solutions: The detailed configuration data for the generated Pareto front. This includes specific decision variables (number of shelves, rows, layers, shuttles, and lifts) and their corresponding objective values (system construction costs and expected order completion cycles) for different scheme types (cost-priority, efficiency-priority, and balanced configurations).This data validates the effectiveness of the proposed RL-IMOEA/D algorithm in solving complex discrete optimization problems and provides a reference for the initial configuration of high-density automated warehousing equipment.
Source: ScienceDB
Institution: Taiyuan University of Science and Technology
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7171531936
