In this study, the spatiotemporal distributions of highway blockage and the low-visibility weather events in eastern China are studied by taking Jiangsu Province as an example. Based on the record table data of highway-blocking events, a vulnerability evaluation model for the highway network in Jiangsu Province is established using the weight assignment methods of the fuzzy analytic hierarchy process (FAHP) and criteria importance though intercriteria correlation (CRITIC). By using the geographic information system, the vulnerability evaluation map of road network in low-visibility weather in Jiangsu Province is finally drawn. The results show that the monthly blockage events on Jiangsu highways are more frequent in the north than in the south and are more frequent along the coast than inland, with the highest occurrence number in winter and a second peak in May. There are basically no blockage events from July to October. Traffic blockage on Jiangsu highways mainly occurs between 22:00 and 08:00 Beijing time. In the afternoon, there are almost no highway-blocking events caused by low-visibility weather. The vulnerability of highway blockage in Jiangsu Province is high in the north and low in the south and high in coastal areas and relatively low in inland. The section K6-K99 of the G30 Lianhuo Highway is the most sensitive.
The durability of road infrastructures strongly correlates with thermal variations. This study proposed a theory-guided multi-scale temporal fusion network (MSTF-Net), integrating multi-scale temporal feature pyramids, dual-pooling mechanisms, and real-time spatiotemporal attention mechanisms for road surface temperature (RST) nowcasting. Through comprehensive evaluations, MSTF-Net achieved 0.66°C mean absolute error with 0.01 ± 0.01°C uncertainty and 74% and 92% threat score for hot and cold extremes respectively, significantly outperforming Multiple linear regression (Linear), Random Forest (RF), eXtreme Gradient Boosting (XGB) and Long Short-term Memory (LSTM) (p < 0.05) in RST prediction accuracy, spatial generalization, weather adaptability and extreme event detections. Systematic ablation studies revealed a feature extraction hierarchy of maximum pooling in micro branches>average pooling in micro branches>real-time spatiotemporal attention mechanisms>macro branches. MSTF-Net combined lightweight computational characteristics (0.11 MB) with high-resolution modeling capabilities, enabling efficient real-time emergency response optimization (38.46 FPS), potentially providing new capabilities in infrastructure-resilient smart systems for thermal risk identification and road predictive maintenance.