Multi-AGV dispatching and routing problem based on a three-stage decomposition method

Automatic guided vehicle (AGV) is a device for horizontal transportation between quay cranes and yard cranes in an automated container terminal. In which dispatching and routing problem (DRP) of the AGV system is a vital as well as basic issue. In the application of the actual AGV system, several practical factors including avoiding conflicts, path smoothness, difficulty in adjusting routes and anti-interference must be considered. The present study establishes the model with the goal of minimizing AGV travel distance, reducing operation time and response time. Furthermore, a three-stage decomposition solution to the problem was proposed by combining the advantages of pre-planning algorithm and real-time planning algorithm, which combines A^* algorithm with the principle of time window to plan the path of each AGV in time order. Finally, the effectiveness of this method in path search and time optimization is illustrated and the system efficiency is improved by comparing and analyzing the calculation examples of different scales.