The scheduling of harbor tugboats is a crucial task in port operations, aiming to optimize resource allocation and reduce operational costs, including fuel consumption of tugboats and the time cost of vessels waiting for operations. Due to the complexity of the port environment, traditional scheduling methods, often based on experience and practice, lack scientific and systematic decision support, making it difficult to cope with real-time changes in vessel dynamics and environmental factors. This often leads to scheduling delays and resource waste. To address this issue, this study proposes a mathematical model based on fuzzy programming, accounting for the uncertainty of the arrival time of target vessels. Additionally, we introduce the NRPER-DDPG algorithm (DDPG Algorithm with Prioritized Experience Replay and Noise Reduction), which combines a prioritized replay mechanism with a decaying noise strategy based on the DDPG algorithm. This approach optimizes the time for tugboats to reach the task location as a continuous action space, aiming to minimize the total system cost and improve scheduling efficiency. To verify the effectiveness of the mathematical model and algorithm, this study conducted experimental validation. Firstly, the optimal algorithm hyperparameter combinations were adjusted through random examples to ensure the stability and reliability of the algorithm. Subsequently, large-scale examples and actual port cases were used to further verify the performance advantages of the algorithm in practical applications. Experimental results demonstrate that the proposed mathematical model and algorithm significantly reduce system costs and improve scheduling efficiency, providing new insights and methods for the sustainable development of port operations.

Economic evaluation of Vessel Traffic Service (VTS) has long relied on sampling-based observations and average-cost assumptions, yet limited attention has been paid to whether evaluation frameworks themselves shape how the value of digital maritime governance is identified and represented. Drawing on full-sample vessel trajectories and dispatch records from a major Chinese hub port during 2018–2020, this study develops a parallel comparative design to compare a traditional sampling-based framework with a full-sample evaluation approach under identical traffic conditions. The results show that differences in evaluation frameworks affect not only the overall magnitude of estimated benefits, but also their composition. Compared with the traditional approach, the full-sample framework generates substantially higher estimates of economic value and assigns greater importance to routine coordination, where repeated small operational adjustments accumulate over time. By contrast, weather disruptions and channel-constrained coordination become relatively less prominent. These differences appear to be associated with variation in temporal benchmarks, sample coverage, and cost assignment strategies. Rather than identifying a single “correct” valuation, the study highlights the sensitivity of VTS economic assessment to methodological assumptions and suggests that, in digitally coordinated public infrastructure systems, evaluation frameworks may shape which forms of governance value become visible and incorporated into policy and investment decisions.

Container terminals constitute a critical component of port logistics systems, and their operational efficiency directly affects cargo turnover and terminal resource utilization. With the continuous growth in container throughput and terminal operating scale, coordination among quay cranes, yard cranes, and internal trucks has become increasingly complex, while equipment waiting and prolonged operation times constrain terminal efficiency. Therefore, efficient coordinated scheduling of yard cranes and internal trucks under quay crane operation sequence constraints is essential for improving terminal operational efficiency and resource utilization. This paper investigates the coordinated scheduling of yard cranes and internal trucks driven by the operational sequence of quay cranes in container terminal handling systems. To address the multi-objective optimization of minimizing yard crane operation time and internal truck waiting time, a mixed-integer linear programming model is developed, considering quay crane operation sequences, yard crane movement paths, and internal truck waiting times. The NoisyNet-A3C algorithm is introduced by incorporating learnable randomness into the parameter space to enhance adaptive exploration and prevent premature policy convergence. The proposed model and algorithm are validated using simulation cases and 30 consecutive days of operational data from Haitian Terminal at Xiamen Port. Simulation results demonstrate that the proposed method outperforms other deep reinforcement learning algorithms. The case study further shows that, compared with the terminal’s existing scheduling scheme, the proposed method reduces yard crane operation time by an average of 23.9% and internal truck waiting time by 21.7%, thereby reducing resource consumption in terminal operations. The results demonstrate that data-driven intelligent coordinated scheduling can support the digital transformation and operational energy optimization of port operations by improving equipment utilization and reducing unnecessary operations and waiting, while providing a practical pathway toward decarbonization and sustainable development of container terminals and shipping.