Digital twin enhanced rescheduling based on hybrid strategy in intermodal container terminal

Abstract

Enhancing the operational efficiency of the railway center station (RCS) is a critical task for intermodal container terminals. Unlike most existing studies that focus on a single container flow, this study investigates the cooperative scheduling of internal container trucks (ICTs) and railway cranes (RCs) for multiple container flows. A mixed-integer programming model is developed to minimize the maximum completion time, the waiting time of external container trucks (ECTs), and the transportation time of ICTs. To generate an optimized baseline schedule, a Multi-Objective Particle Swarm Optimization (MOPSO) algorithm is employed. From a practical operational perspective, however, the baseline schedule must effectively address uncertainties in real-time. Therefore, we propose a proactive-reactive and global-local hybrid rescheduling strategy based on digital twin (DT). The DT facilitates uncertainty monitoring, rescheduling plan generation, simulation, and iterative optimization. The hybrid strategy consists of: (1) a global rescheduling optimization model, which is proactively triggered periodically or reactively triggered due to cumulative errors, aiming to maximize schedule stability; and (2) a local short-interval recovery policy, which is reactively triggered to handle uncertainties occurring between two consecutive global rescheduling points. A case study is conducted to demonstrate the effectiveness of the proposed rescheduling strategy in handling delayed ECT arrivals and other uncertainties. The results highlight the efficiency of the DT-enhanced methodology in improving RC and ICT collaboration. Sensitivity analysis further identifies appropriate threshold settings and equipment configurations. The results show that the application of the DT-enhanced rescheduling methodology in the RCS helps operators to make optimized and timely decisions.