The Multi-load Automated Guided Vehicle (M-AGV) has emerged as a key enabling technology for intelligent and sustainable workshop logistics owing to its potential to enhance transportation efficiency and reduce system costs. To address the limitations in energy optimization caused by simplified AGV speed and payload modeling in existing scheduling models, this study develops a multi-factor coupled energy consumption model—integrating vehicle speed, travel distance, and dynamic payload—to minimize the total energy consumption of M-AGV systems. To effectively solve the model, a hybrid optimization algorithm that combines the State–Action–Reward–State–Action (SARSA) learning algorithm with the Triangulation Topology Aggregation Optimizer (TTAO), complemented by a similarity-based individual generation strategy, is designed to jointly enhance the algorithm’s exploration and exploitation capabilities. Comparative experiments were conducted across task scenarios involving three different handling task scales and three levels of M-AGV fleet heterogeneity, demonstrating that the proposed SARSA-TTAO algorithm outperforms Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and the Hybrid Genetic Algorithm with Large Neighborhood Search (GA-LNS) in terms of solution accuracy and convergence performance. The study also reveals the differences between homogeneous and heterogeneous M-AGV fleets in task allocation and resource utilization under energy-optimal conditions.

The increasing demand for storing various types of goods has led to a raise in the need for storage capacity in warehousing systems. Autonomous vehicle storage and retrieval systems (AVS/RSs) offer high flexibility by allowing different configurations to meet different storage requirements. The system mainly completes operations through elevators and multiple rail-guided vehicles (RGVs). This paper focuses on the scheduling optimization of compound operations in the AVS/RS to improve system performance. Compound operations involve the coordinated execution of both single-command and double-command operations. A mathematical model with compound operations was proposed and effectively decomposed into a horizontal component for RGVs and a vertical counterpart for the elevator, which can represent the operations of one elevator cooperating with multiple RGVs. The goal of this model was to minimize the makespan for compound operations and to determine the optimal operation sequence and path for RGVs. An improved discrete particle swarm optimization (DPSO) algorithm called AGDPSO was proposed to solve the model. The algorithm combines DPSO and a genetic algorithm in an adaptive manner to prevent the algorithm from falling into local optima and relying solely on the initial solution. Through rigorous optimization, optimal parameters for the algorithm were identified. When assessing the performance of our improved algorithm against various counterparts, considering different task durations and racking configurations, our results showed that AGDPSO outperformed the alternatives, proving its effectiveness in enhancing system efficiency for the model. The findings of this study not only contribute to the optimization of AVS/RS but also offer valuable insights for designing more efficient warehouses. By streamlining scheduling, improving operations, and leveraging advanced optimization techniques, we can create a more robust and effective storage and retrieval system.