The efficiency and low-carbon operation of automated container terminals (ACTs) are closely related to the coordinated scheduling of quay cranes (QCs), yard cranes (YCs), and automated guided vehicles (AGVs). Under mixed import and export tasks, unreasonable equipment allocation often leads to prolonged AGV waiting and unnecessary energy consumption. This study focuses on the green AGV scheduling problem with AGV-mate capacity constraints and establishes a mixed-integer programming model aimed at minimizing the total energy consumption of AGVs. The model incorporates energy consumption characteristics of AGVs during loaded travel, empty travel, and waiting periods. A learning-enhanced adaptive large neighborhood search algorithm (L-ALNS) is developed with an ε-greedy operator selection mechanism and dynamic weight update strategy to improve search efficiency and solution quality. Numerical results demonstrate that L-ALNS outperforms benchmark algorithms in total system energy consumption under various scales. Further analysis reveals that AGV and AGV-mate configurations significantly affect system performance, while excessive resource deployment may lead to diminishing marginal benefits. Overall system performance depends on the dynamic balance among horizontal transport capacity, yard-side buffer capacity, and equipment coordination efficiency.
Zhang et al. (Fri,) studied this question.