Multi-Agent Path Planning for a Multi-Deep Four-Way Shuttle-Based System

Background : Four-way shuttle-based storage and retrieval systems (FSS/RSs) have recently emerged as flexible and scalable solutions for high-density warehousing, enabling shuttle movement in four directions and supporting multi-deep dual-access storage configurations. However, these features increase the complexity of vehicle coordination and collision management. This study proposes a multi-agent path-planning methodology for FSS/RSs with multi-deep dual-access lanes hosting homogeneous items. Methods : An A*-based path-planning framework was developed and integrated with a dynamic collision-management strategy comprising collision detection, priority assignment, and collision avoidance. The methodology was evaluated through a multi-scenario analysis considering different fleet sizes, priority-assignment strategies, safety-area extensions, transaction-entry patterns, and collision-management policies. Results : The results show that fleet size is the most influential operational parameter, significantly affecting throughput, waiting times, and collision frequency. Increasing the number of vehicles improves system productivity but also increases traffic interactions and congestion. The analyses further highlight the effects of safety-area size, priority rules, and transaction-entry patterns on operational performance and system robustness. Conclusions : The proposed methodology effectively combines path planning and collision management in four-way shuttle systems, providing a decision-support tool for evaluating operational trade-offs among throughput, congestion control, and system stability in multi-deep dual-access warehouse environments.