Multimodal transport is a process of effectively moving cargoes in a single container by combining land transport (road or rail) and maritime or river transport (vessel or barge) in one transport chain. However, cold chain logistics (CCL), as a special while major kind of cargo delivery, has not been incorporated with this beneficial combination. In order to realize efficient delivery of cold chain foods (CCF), in this study, the characteristics of multimodal and CCL are analyzed and integrated to select the optimal logistics path. In establishing the path-selection model, customer satisfaction is introduced, which is reflected by arrival punctuality and the quality of CCF. An improved particle swarm optimization algorithm (IPSO) is introduced to address the model and is proven to retain a fast convergence rate and achieve outstanding solving accuracy through the experimental study. Sensitivity analysis is also conducted to present the impact of railway speed and cost variation on path selection. Results show that compared with highway transport, railway transport is preferable to the medium and long distance. The influence of railway speed improvement is more striking than cost reduction in motivating decision makers to choose railway transport mode in logistics operations.

As an affordable form of public transit, the metro still faces first- and last-mile problems. Integration of dockless bike sharing (DBS) and the metro offers a reasonable solution to this problem. However, an in-depth exploration of the spatio-temporal characteristics and determinants of DBS–metro integration, especially meteorological indicators, is lacking. The spatiotemporal characteristics of the transfer volume on weekdays and weekends in Beijing were analyzed. Geographically weighted regression was utilized to investigate the effects of the significant determinants on the spatial variation of the transfer volume. The results indicate that the transfer volume is concentrated within the Fifth Ring Road in Beijing. The effects of education services and metro station density were negatively correlated with transfer volume. As expected, population density and enterprises positively affect transfer volume, whereas wind speed and relative humidity have negative effects. These findings provide new insights into the promotion of DBS–metro integration.

In recent years, the frequency of sudden disaster events has increased significantly, leading to substantial and often irreparable losses for society and individuals. Although the comprehensive development of China’s transportation network has laid a solid foundation for the transportation of emergency supplies, the construction of an optimized transportation system for emergency logistics still faces considerable challenges, particularly given the uncertainties of the transportation environment. To ensure the efficient delivery of emergency supplies to disaster sites via optimal routes and the utilization of multimodal transport for logistics distribution, this paper addresses the following key objectives. First, recognizing the uncertainties in parameters such as transportation time and cost due to dynamic changes in the transportation environment as well as the limitations of transshipment capacity, we developed a multimodal transport path optimization model for emergency supplies. The model aims to minimize both transportation time and costs. Second, in addressing uncertain parameters, we employed the theory of interval number ranking, defining a risk coefficient to convert uncertain parameters into deterministic ones. This enabled a deterministic transformation of the model. Additionally, a weighted scoring method was applied to assign weights to each objective, effectively transforming the multi-objective optimization problem into a single-objective problem. Finally, we applied the genetic algorithm, particle swarm optimization algorithm, and a hybrid genetic–particle swarm algorithm to solve the model. The effectiveness of the model and the superiority of the hybrid algorithm were verified through case studies. Furthermore, a sensitivity analysis of the parameters and an investigation into the influence of the risk coefficient were conducted, providing valuable insights for relevant decision-making departments in formulating emergency material transportation plans.

The paper presents an improved iterative prediction method for bus arrival time at multiple downstream stops. A multiple-stop prediction model includes two stages. At the first stage, an iterative prediction model is developed, which includes a single stop prediction model for arrival time at the immediate downstream stop and an average bus speed prediction model on further segments. The two prediction models are constructed with a support vector machine (SVM). At the second stage, a dynamic algorithm based on the Kalman filter is developed to enhance prediction accuracy. The proposed model is assessed with reference to data collected on transit route No 23 in Dalian city, China. The obtained results show that the improved iterative prediction model seems to be a powerful tool for predicting multiple stop arrival time.

The key to intelligent traffic control and guidance lies in accurate prediction of traffic flow. Since traffic flow data is nonlinear, complex, and dynamic, in order to overcome these issues, graph neural network techniques are employed to address these challenges. For this reason, we propose a deep-learning architecture called AMGC-AT and apply it to a real passenger flow dataset of the Hangzhou metro for evaluation. Based on a priori knowledge, we set up multi-view graphs to express the static feature similarity of each station in the metro network, such as geographic location and zone function, which are then input to the multi-graph neural network with the goal of extracting and aggregating features in order to realize the complex spatial dependence of each station’s passenger flow. Furthermore, based on periodic features of historical traffic flows, we categorize the flow data into three time patterns. Specifically, we propose two different self-attention mechanisms to fuse high-order spatiotemporal features of traffic flow. The final step is to integrate the two modules and obtain the output results using a gated convolution and a fully connected neural network. The experimental results show that the proposed model has better performance than eight other baseline models at 10 min, 15 min and 30 min time intervals.