With the rapid development of world economy and trade and the continuous construction of green port, automated container terminal (ACT) has increasingly become the direction of future development. Layout design is the premise of ACT construction, which has an at least 50-year influence on the terminal. Therefore, this paper hopes to analyze and design the typical ACT layout to achieve sustainable development of the port. Firstly, a conceptual model is presented considering the interaction between different areas within the ACT when the width and length of the terminal are fixed. To select the optimal layout to achieve the goal of the green terminal, a novel mathematical model is established based on the energy consumption during cycle operation of various devices which can estimate the total carbon emission of an ACT over a period and is suitable for designing period. Then, with the developed model, an ACT in East China was taken as a case study. Finally, according to various analysis of the data results, the layout suggestion considering the sustainable development of the port is given.

The dispatching of automated guided vehicles (AGVs) is essential for efficient horizontal transportation at automated container terminals. Effective planning of AGV transportation can reduce equipment energy consumption and shorten task completion time. Multiple AGVs transport containers between storage blocks and vessels, which can be regarded as the supply sides and demand points of containers. To meet the requirements of shipment in terms of timely and high-efficient delivery, multiple AGVs should be dispatched to deliver containers, which includes assigning tasks and selecting paths. A contract net protocol (CNP) is employed for task assignment in a multiagent system, while machine learning provides a logical alternative, such as Q-learning (QL), for complex path planning. In this study, mathematical models for multi-AGV dispatching are established, and a QL-CNP algorithm is proposed to tackle the multi-AGV dispatching problem (MADP). The distribution of traffic load is balanced for multiple AGVs performing tasks in the road network. The proposed model is validated using a Gurobi solver with a small experiment. Then, QL-CNP is used to conduct experiments with different sizes. The other algorithms, including Dijkstra, GA, and PSO, are also compared with the QL-CNP algorithm. The experimental results demonstrate the superiority of the proposed QL-CNP when addressing the MADP.

This study proposes a digital twin (DT) application framework that integrates deep reinforcement learning (DRL) algorithms for the dynamic scheduling of crane transportation in workshops. DT is used to construct the connection between the workshop service system, logical simulation environment, 3D visualization model and physical workshop, and DRL is used to support the core decision in scheduling. First, the dynamic scheduling problem of crane transportation is constructed as a Markov decision process (MDP), and the corresponding double deep Q-network (DDQN) is designed to interact with the logic simulation environment to complete the offline training of the algorithm. Second, the trained DDQN is embedded into the DT framework, and then connected with the physical workshop and the workshop service system to realize online dynamic crane scheduling based on the real-time states of the workshop. Finally, case studies of crane scheduling under dynamic job arrival and equipment failure scenarios are presented to demonstrate the effectiveness of the proposed framework. The numerical analysis shows that the proposed method is superior to the traditional dynamic scheduling method, and it is also suitable for large-scale problems.

With the continuous development of smart ports, the construction of port information systems has become the construction focus of the future development of such ports. However, the existing performance evaluation system for port information systems represents a research gap. It cannot effectively achieve specific improvement in the construction of information systems for port services and capabilities. In this paper, an evaluation index system was forged from the three perspectives of the operating level, management capability, and economic efficiency. As a first step, starting from the current situation of Beibu Gulf Port (Port B) construction, the three dimensions of the evaluation index system were determined using the scheme of literature review, questionnaire research, and fieldwork. Additionally, through adversarial interpretive structural modeling (AISM), it is concluded that the port loading and unloading operation capability and service level are the fundamental factors for measuring the construction performance of the container terminal operation system (CTOS). The results are used as input to construct an analytic network process (ANP) model to obtain the index weights. Finally, the gray clustering method (GCM) is introduced to construct a quantitative evaluation model to quantitatively evaluate the construction performance of the Port B CTOS. The quantitative benefits brought by the construction of the CTOS are finally verified. The scores of the corresponding dimensions of the port before and after the construction of the CTOS system in Beibu Gulf Port are obtained experimentally. The conclusion shows that the construction of CTOS makes the port operating level module improve the most, from 40.023 to 70.733 points; at the same time, it is found that two aspects, i.e., green security and economic benefits, in current construction are directions requiring further work in future port construction. Finally, a visualized quantitative analysis and evaluation method for the performance of smart port construction is proposed.

In this paper, the essence and optimization objectives of the hull parts path optimization problem of CNC laser cutting are described, and the shortcomings of the existing optimization methods are pointed out. Based on the optimization problem of the hull parts CNC laser cutting path, a new part-cutting constraint rule based on partial cutting is proposed, which aims to overcome the drawbacks of the traditional algorithms with serial cutting constraint rules. This paper addresses the problem of optimizing the path for CNC laser cutting of hull parts, including an empty path and the order and directions used for the provided cut contours. Based on the discretization of the part contour segments, a novel toolpath model for hull parts called hull parts cutting path optimization problems based on partial cutting rules (HPCPO) is proposed in this paper. To solve the HPCPO problem, a segmented genetic algorithm based on reinforcement learning (RLSGA) is proposed. In RLSGA, the population is viewed as an intelligent agent, and the agent’s state is the population’s diversity coefficient. Three different segmented crossover operators are considered as the agent’s actions, and the agent’s reward is related to the changes in the population’s fitness and diversity coefficients. Two benchmark problems for HPCPO were constructed to evaluate the performance of RLSGA and compared with four other algorithms. The results showed that RLSGA outperformed the other algorithms and effectively solved the HPCPO problem.

Accurate sales forecasting can provide a scientific basis for the management decisions of enterprises. We proposed the xDeepFM-LSTM combined forecasting model for the characteristics of sales data of apparel retail enterprises. We first used the Extreme Deep Factorization Machine (xDeepFM) model to explore the correlation between the sales influencing features as much as possible, and then modeled the sales prediction. Next, we used the Long Short-Term Memory (LSTM) model for residual correction to improve the accuracy of the prediction model. We then designed and implemented comparison experiments between the combined xDeepFM-LSTM forecasting model and other forecasting models. The experimental results show that the forecasting performance of xDeepFM-LSTM is significantly better than other forecasting models. Compared with the xDeepFM forecasting model, the combined forecasting model has a higher optimization rate, which provides a scientific basis for apparel companies to make adjustments to adjust their demand plans.