The risk of ship–bridge collisions should be evaluated using advanced models to consider different anti-collision and bridge-protection measures. This study aimed to propose a method to evaluate the effectiveness of active and passive safety measures in preventing ship–bridge collision. A novel ship–bridge collision probability formulation taking into consideration different safety measures was proposed. The model was applied at Jintang Bridge in China where the surrounding vessel traffic is ultra-crowded. We calculated the collision probability between the bridge and passing traffic using automatic identification system (AIS) data, Monte Carlo simulation, and Bayesian networks. Results under four different safety measures (i.e., active measures, passive measures, both measures and none) were analyzed and compared. The analysis concluded that both active and passive safety measures are effective in reducing the ship–bridge collision probability. Active measures, if deployed properly, can provide protection at an equivalent level than passive measures against collision risks. However, passive measures, such as setting arresting cables, are necessary in cases where the response time of the active measures is long. The proposed method and the results obtained from the case study may be useful for robust and systematic effectiveness evaluation of safety measures in other cases worldwide.

Taking cold-chain logistics as the research background and combining with the overall optimisation of logistics distribution networks, we develop two-stage distribution location-routing model with the minimum total cost as the objective function and varying vehicle capacity in different delivery stages. A hybrid genetic algorithm is designed based on coupling and collaboration of the two-stage routing and transfer stations. The validity and feasibility of the model and algorithm are verified by conducting a randomly generated test. The optimal solutions for different objective functions of two-stage distribution location-routing are compared and analysed. Results turn out that for different distribution objectives, different distribution schemes should be employed. Finally, we compare the two-stage distribution location-routing to single-stage vehicle routing problems. It is found that a two-stage distribution location-routing system is feasible and effective for the cold-chain logistics network, and can decrease distribution cost for cold-chain logistics enterprises.

Many governments have been strengthening the construction of hardware facilities and equipment to prevent and control marine oil spills. However, in order to deal with large-scale marine oil spills more efficiently, emergency materials dispatching algorithm still needs further optimization. The present study presents a methodology for emergency materials dispatching optimization based on four steps, combined with the construction of Chinese oil spill response capacity. First, the present emergency response procedure for large-scale marine oil spills should be analyzed. Second, in accordance with different grade accidents, the demands of all kinds of emergency materials are replaced by an equivalent volume that can unify the units. Third, constraint conditions of the emergency materials dispatching optimization model should be presented, and the objective function of the model should be postulated with the purpose of minimizing the largest sailing time of all oil spill emergency disposal vessels, and the difference in sailing time among vessels that belong to the same emergency materials collection and distribution point. Finally, the present study applies a toolbox and optimization solver to optimize the emergency materials dispatching problem. A calculation example is presented, highlighting the sensibility of the results at different grades of oil spills. The present research would be helpful for emergency managers in tackling an efficient materials dispatching scheme, while considering the integrated emergency response procedure.

Abnormal ship navigation behaviors in cross-sea bridge waters pose significant threats to maritime safety, creating a critical need for accurate anomaly detection methods. Ship AIS trajectory data contain complex temporal features but often lack explicit labels. Most existing anomaly detection methods heavily rely on labeled or semi-supervised data, thus limiting their applicability in scenarios involving completely unlabeled ship trajectory data. Furthermore, these methods struggle to capture long-term temporal dependencies inherent in trajectory data. To address these limitations, this study proposes an unsupervised trajectory anomaly detection model combining a transformer architecture with a variational autoencoder (transformer–VAE). By training on large volumes of unlabeled normal trajectory data, the transformer–VAE employs a multi-head self-attention mechanism to model both local and global temporal relationships within the latent feature space. This approach significantly enhances the model’s ability to learn and reconstruct normal trajectory patterns, with reconstruction errors serving as the criterion for anomaly detection. Experimental results show that the transformer–VAE outperforms conventional VAE and LSTM–VAE in reconstruction accuracy and achieves better detection balance and robustness compared to LSTM–-VAE and transformer–GAN in anomaly detection. The model effectively identifies abnormal behaviors such as sudden changes in speed, heading, and trajectory deviation under fully unsupervised conditions. Preliminary experiments using the POT method validate the feasibility of dynamic thresholding, enhancing the model’s adaptability in complex maritime environments. Overall, the proposed approach enables early identification and proactive warning of potential risks, contributing to improved maritime traffic safety.