Improving Container Port Efficiency: A Data-Driven Model for Optimizing Truck Arrival Appointments Through Distributionally Robust Optimization

Abstract

The irregular arrival patterns of container trucks at ports have a substantial impact on logistics operations’ efficiency, resulting in congestion during peak hours and unused port capacity during idle times. Implementing a truck appointment system (TAS) is vital to address this issue effectively. This paper suggests enhancing the TAS by adopting a data-driven approach using terminal gate data to understand the intricate and uncertain relationship between truck arrival patterns and port operational efficiency. Insights gained from these data are utilized to develop a distributionally robust optimization (DRO) model. This model provides an exact solution for optimizing the appointment quota plan of TASs, thereby improving port efficiency and addressing operational challenges. Compared to existing methods, this approach does not heavily rely on theoretical assumptions concerning the cooperation mechanisms among trucks, yard equipment, quayside equipment, and other facilities and fully considers the complex uncertainties in truck arrivals. Furthermore, to examine the effectiveness of the proposed model, a case study is conducted at Yan Port, China, aiming to achieve practical results. The numerical experiments comparing its performance with the conventional robust optimization (RO) model confirm the superiority of the proposed DRO model in minimizing the total truck turnaround time within the terminal and overall time expenses. This superiority stems from its integration of the respective advantages of stochastic optimization (SO) and traditional RO methods. By optimizing the appointment quota plan in this manner, it achieves a balanced distribution of truck arrivals, showcasing its significant potential to enhance port logistics efficiency.