In order to clarify the influence of droop angle on suspended cable fire spread characteristics, a cable combustion test platform was constructed. The cable combustion tests with different droop angles (θ) and different cross-sectional areas (S) were carried out. The results show that the flame presents an axisymmetric conical state when the cable is arranged horizontally. When the droop angle is greater than 0°, as the combustion progresses, the flame shape transits from a multi-branched cone to an extended state along the cable direction. The flame spread behavior of cable can be divided into three stages: the advance of the pyrolysis front in stage I increases FSR, the interaction between thermal convection and thermal radiation in stage II leads to a rapid increase of FSR. Stage III thermal radiation dominates the continuous increase of FSR. The radiation heat flux generated by combustion plays a leading role in the increase of FSR. The rapid increase of radiation heat flux and the appearance of peak value are consistent with the three stages of fire spread, which can be used as a sign of the increase stage. The increase of θ and the decrease of S will promote the increase of FSR. According to the Stephen-Boltzmann law, the relationship between θ and S for the increase of FSR is proposed, and the theoretical index of the relationship is in good agreement with the experimental index.

In order to increase the systematicity and flexibility of disaster response behavior at the grassroots level, a triangle model of county disaster resilience was constructed. Taking the rainstorm in Jianchang, Liaoning Province in 2024 as an example, social network analysis method was used to establish an association network among rainstorm disaster events, county disaster bearing bodies and emergency management plans. The system dynamics model reflecting the evolution process of county disaster resilience was designed by introducing the time series dimension, and the individual attributes of network nodes were used as the basis for model evaluation. The results show that the county disaster resilience shows an evolutionary trend of slowly decreasing first and then steadily increasing, and the growth rate gradually accelerates. During the pre-disaster warning stage, disaster events, disaster-bearing bodies, and emergency management plans respectively possess the characteristics of suddenness, exposure, and absorption. Organizational preparations, communication preparations, material preparations, and publicity preparations have the greatest room for optimization. During the disaster response stage, disaster events, disaster-bearing bodies, and emergency management plans respectively possess coupling and uncertainty, sensitivity, and adaptability characteristics. There is still room for improvement in the capacity to evacuate the masses, communication guarantee mechanisms, and the resilience of power facilities. During the post-disaster recovery stage, disaster-bearing bodies and emergency management plans respectively possess adaptability and adjustment characteristics. At this time, it is necessary to optimize the planting structure of crops and enhance the flood resistance capacity of roads. Emergency management plans always have the characteristic of learning throughout the entire cycle of disaster response.

To optimize vehicle design and improve road safety, the effect of the NBLEH on pedestrians head injuries during ground contact was investigated. The impact with the ground AIS 1+ pedestrian landing injury cases was selected, and data analysis along with multicollinearity tests was conducted. The results show that as the NBLEH increases, the severity of pedestrian head injuries exhibits a corresponding upward trend. Through Logistic regression analysis, it is further revealed that for AIS 2+ head injuries in adult pedestrians can be jointly predicted by vehicle speed, age, and NBLEH. Specifically, the risk of injuries is increased with higher values of NBLEH, faster speeds, and older ages. Analysis of real accident videos demonstrates that a smaller NBLEH results in increased full-body rotation angles of pedestrians. However, once NBLEH surpasses 0.9, the rotation angle notably diminishes. The collision force transmission and the pedestrian's landing posture are altered by different values of NBLEH. Additionally, cadaver crash tests were conducted using two types of vehicles with different front-end heights to establish test conditions with different NBLEH values. The results indicate that NBLEH not only influences the timing of head contact with the ground, but also that there is an interaction between head landing posture and collision speed, further complicating head injuries. Parametric simulation studies reinforced that NBLEH significantly influences factors such as posture, angle, timing, and speed during pedestrian head-to-ground contact, thereby increasing the complexity and uncertainty of surrounding injury mechanisms. The simulation results show that both vehicle speed and NBLEH influence the pedestrian flip angle, with NBLEH values less than 1 producing larger pedestrian flip angles than values greater than 1.

In order to study the influence characteristics of different driving conditions on driving behavior of small-clearance interchanges, eye tracker was used to collect the drivers' eye movement data in the real vehicle experiment. k-medoids dynamic clustering method was used to divide the drivers' visual interest area. A comparative analysis of gaze behavior and saccade characteristics of drivers was conducted of under mainline entry and exit conditions at conventional clear-distance and small-clearance interchanges and across different driving stages. The driving behavior characteristics of drivers under high-density interchange of the expressway were revealed. The results show that drivers mainly focus on the area of road ahead and instrument panel, and they also pay special attention to the area of the left mirror when driving into mainline conditions. The 500 ms gaze time is a common time window for drivers to capture key information. The mean value of the gaze time on small-clearance interchange is higher than that on regular-clearance interchanges. There is a clear bias in observing the horizontal and vertical information in different driving phases. The mean gaze duration was higher in the small-clearance interchanges than in the conventional interchanges. The reduction of clear distance of the interchanges increases drivers' scanning amplitude and changed the scanning direction, which is most obvious in the interweaving section. The scanning amplitude is uniformly distributed in the range of 200 px. There is no significant difference in the mean distribution of fixation time between drivers of different driving ages. There is no significant monotonic relationship between the drivers' gaze duration and scanning speed and amplitude, while there is a significant positive correlation between scanning speed and amplitude of scanning.

To enhance urban resilience against disaster chains, this study proposed an analytical framework and methodology for identifying influencing factors from a disaster chain perspective. Firstly, a comprehensive list of urban resilience influencing factors was established by focusing on the core capacities of resilient cities within disaster chain contexts. Secondly, an improved fuzzy DEMATEL-ISM integrated method was introduced to identify top-level factors, clarify hierarchical structures and interaction pathways among influencing factors, and analyze resilience enhancement approaches from economic and demographic perspectives. Finally, taking the resilience construction practice of Chengdu in the context of disaster chain as an example, the rationality of this theoretical analysis was verified. Results indicate that the top-level design factors, aimed at addressing disaster chains, primarily focus on prevention, preparedness, and adaptation. Fundamental factors include insurance protection, emergency command in the disaster chain scenario and government departments' ability to jointly handle chain disruptions to reduce disasters, adequacy of emergency resource preparation, and ecological environment restoration and governance. Direct influencing factors encompass housing quality, medical assistance capacity, emergency evacuation capacity, and basic living security. The most effective approaches to enhance economic resilience in disaster chains are strengthening insurance protection and accelerating infrastructure repair, both promoting economic recovery. Enhancing population resilience relies primarily on measures such as housing security, life support, emergency evacuation support, and basic living security. Chengdu maintains relatively sound resilience construction under disaster chain scenarios, though certain aspects require further optimization.

To optimize the emergency management system for plateau disasters and to address the scientific demand for improving governance effectiveness under extreme plateau environments, this study developed a dual-network framework—"risk propagation-emergency response"—using the Ms6.8 Shigatse earthquake in Tibet as a case study. Based on complex network theory, it uncovered the cascading dynamics of seismic risks and multi-actor coordination governance patterns. The analysis demonstrates that the cascading disaster network exhibits a core-periphery structure, where critical nodes trigger multi-level chain reactions via infrastructure fragility and cultural sensitivity. Plateau-specific disaster chains are identified, such as "cold survival crisis → mass livestock deaths → pressure of harmless treatment", alongside culturally embedded risks like "language barriers" and "religious site damage", and their impacts further promote the cultural adaptive reconstruction of the traditional emergency management system. The coordination network shows a decentralized, short-path, high-density topology, indicating efficient collaboration despite persistent hierarchical redundancy.

In order to study the scientific and reasonable location selection of regional emergency rescue stations and base-level emergency rescue stations in the planning and construction stage of the subway network, firstly, the topology model of the subway network was constructed based on complex network theory. Relevant indicators were selected to construct the evaluation system of node importance, and GWR model was used to predict passenger flow to supplement the research data. Secondly, aiming at the highest efficiency, optimal coverage and lowest cost, the multi-objective location model of regional emergency rescue stations was established, and NSGA-Ⅱ algorithm was used to solve the optimal scheme. At the same time, a hybrid analysis calculation method was designed to solve the comprehensive importance value parameters of nodes in the model, and based on the site selection results of regional emergency rescue stations, the enumeration method was used to solve the optimal solution of base-level emergency rescue stations. Finally, Xi'an metro planning network was taken as an example to verify the calculation. The results show that compared with the optimal solution in the case of "no consideration of node importance", the regional location solution with "consideration of node importance" saves about 2.19% in cost, improves the rescue efficiency by about 9.43%, and exceeds the optimal solution with about 4.66% in coverage. Therefore, the location scheme of "considering node importance" is more in line with the actual needs of the subway network.

In order to more accurately capture the evolutionary patterns of disaster risks and facilitate scientific risk assessment, this study innovatively integrated the dual natural and social attributes of disasters. By leveraging the outcomes from event extraction, event relationship identification, and event generalization, a method based on causal association strength was developed to mine event logic. Subsequently, an emergency event evolutionary graph was constructed for assessing chain-reaction disaster risks. Combined with a similarity matching method that incorporates the multi-dimensional attributes of disaster events, an empirical study was conducted using earthquake disasters as a case study. The results demonstrate that the proposed disaster risk assessment framework based on the event evolutionary graph can rapidly identify historical earthquake events with high similarity to the Luding earthquake case. Furthermore, by leveraging the intrinsic interconnections and evolutionary paths among risks revealed by the graph, the framework can predict secondary disasters (e.g., landslide-dammed lakes) and derivative social risks (e.g., rumor propagation).

In order to realize scientific and efficient post-disaster transfer of the population after the occurrence of major natural disasters, the emergency bus evacuation planning considering the transfer urgency was studied according to the different severity of the disaster. Firstly, a continuous emergency bus evacuation model was established with the goal of minimizing the evacuation completion time and evacuation delay satisfaction degree. Secondly, a HS-GA based on transfer urgency coding was designed and applied for problem-solving, and the effectiveness of the algorithm was verified using examples of different scales. Finally, the evacuation planning considering the transfer urgency was compared with those without considering the transfer urgency based on the case. The results show that the HS-GA is better than Genetic Algorithm (GA) and Neighborhood Search GA (NS-GA). According to the case solved by HS-GA, a comparison is made with evacuation planning that does not consider transfer urgency. When transfer urgency is considered, the evacuation delay dissatisfaction degree is reduced by 51.4%. Additionally, shorter evacuation completion times are observed at both Class I and Class II disaster sites. The average time taken to arrive at the shelter is also found to be shorter. At the same time, the dissatisfaction degrees of evacuation delay considering the urgency of transfer under different emergency bus quantity are all lower. The evacuation model considers the evacuation priority of the hard-hit areas under different severity of the disaster, and can make reasonable dispatch arrangements under the circumstance of limited rescue vehicles, which is more objective and in line with the actual situation.

In order to explore the role of intelligent and digital empowerment in the modernization of emergency management systems and capacities, this study identified the key influencing factors and clarifies their interrelationships. CiteSpace was applied to cluster and analyze relevant domestic and international literature, while NVivo12plus, combined with expert consultations, was used to identify 16 influencing factors from four dimensions: environment, mechanism, technology, and data. DEMATEL method was employed to calculate the influence, dependence, centrality, and causality of each factor, and ISM was used to construct a hierarchical structure and reveal their internal logical relationships. By integrating the DEMATEL and ISM models, five key factors were identified, namely political and legal frameworks, cultural awareness, demand and supply, management strategies and planning, and managerial attention and support, which rank highest in multiple indicators and play a significant role in promoting modernization.