Narrow Results

Passenger evacuation time prediction is a crucial basis for subway station management to better formulate safety control measures. It becomes possible to reasonably plan the flow of passengers within the station and configure safety devices such as signage and sprinkler systems with a known and explicit time framework, thereby reducing passenger congestion and panic. However, methods based on crowd dynamics simulation require a significant amount of time and effort to build models, and the rapidity of time prediction is challenging to ensure. Real human evacuation experiments involve ethical, safety and practical operational issues. To address this challenge, an evacuation time prediction model for subway passengers is established based on the CPA-SVR machine learning method, enhancing the speed and accuracy of prediction. The reliability of simulation results is validated by comparing observed values of passenger alighting and boarding time and traffic time at stairs with simulation values from MassMotion software. Fourteen factors related to the subway station structure, passengers and train status are selected as influence factors for evacuation time. A foundation data set for the evacuation time prediction model is obtained through 179 evacuation experiments under different influence factors using the MassMotion simulation system at 32 constructed stations. The SHAP interpretation method is applied to identify the importance of influence factors in the experimental results. A CPA-SVR passenger evacuation time prediction model is established, with accuracy concentrated between 85%–100%, based on training and validation sets. Further testing with 45 additional sets of fresh experimental data demonstrates the model’s strong predictive capability for new data, indicating good generalization ability. Finally, a focused analysis of passenger evacuation behaviors at bottlenecks such as stairs, gates and exits is conducted, accompanied by relevant optimization strategies.

In the present work we studied the room evacuation problem using the social force model introduced by Helbing and coworkers. The "faster is slower" effect induced by panic was analyzed. It could be explained in terms of increasing mean clogging delays which shows a strong correlation with certain structures that we call "blocking clusters". Also a steady state version of the problem was implemented. It shows that, from a macroscopic point of view, the optimal evacuation efficiency correspond to the state at which the difference between the system desire force minus the system granular force is maximum.

In order to characterize the disturbance fluctuation of pedestrian flow caused by the disturbance during evacuation and the state change of pedestrian flow, this paper improves the social force model by introducing disturbance fluctuation force. First, a momentum equation is established to describe the change of pedestrian flow under the influence of disturbance fluctuation, and the mathematical expression of disturbance fluctuation force is given. Second, the evacuation processes of pedestrian flow with and without “queue jumpers” are simulated with the simulation experimental platform, and the key factors influencing the performance of the model are deeply studied through numerical analysis. The results showed that: when the expected velocity is the same, the bigger the angle between the cross-section position vector and the initial expected velocity is, the more serious the congestion occurs at the exit. In addition, when the crowd density is small, the larger the angle, the higher the evacuation efficiency and vice versa.

The traffic environment in tunnel entrance area is complex, with prominent issues related to safety and efficiency. However, existing models find it difficult to accurately capture the traffic situation in this area. Therefore, this paper establishes a vehicle dynamics model for the entrance area of a tunnel based on the social force model and artificial potential field. This model can characterize the deceleration behavior of vehicles at the entrance of the tunnel and the interaction between vehicles. By considering the special effects of the tunnel environment, the simulation experiments successfully reproduce the congestion phenomenon in the entrance of the tunnel. Based on this model, this paper proposes a coordinated driving model under vehicle to vehicle (V2V) communication environment, and the simulation results show that this model can effectively alleviate the congestion in the entrance area of the tunnel and improve traffic efficiency.

Urban flooding events have emerged as increasingly prevalent and severe natural hazards, owing to the influence of climate change and human-induced activities. However, pedestrian locomotion in floodwater diverges significantly from level ambulation due to the complex interaction between pedestrians and water. To systematically investigate this mechanism of interaction, we propose the integration of a drag force into the social force model to simulate pedestrian movement behavior under floodwater conditions. The modified social force model was conducted for sensitivity parameters analysis and calibrated by three controllable experiments. Based on this calibrated model, an in-depth investigation has been conducted to analyze the influence of water depth and water flow velocity on pedestrians’ movement speed. Simulation results suggest that as water depth gradually incrementally rises, the drop rate of speed in running conditions is notably faster than that of walking conditions no matter what water flow speed and direction was. In addition, we propose a mathematical model capable of predicting pedestrians’ movement speed under floodwater conditions. These findings will offer valuable insights into the risk assessment of pedestrian evacuation in flooding scenarios.

There often exist behaviors of moving against the main direction of evacuation in order to rescue or find the important missing people in real situations. However, the traditional social force model (SFM) often lacks consideration of such “counter flow”. Motivated by this, we improve the traditional SFM to study the counter flow and its influence on evacuation out of multi-exit rooms. We call the person to be rescued “superior” and the rescuers “subordinate”. Two different rescue situations are proposed: superior waiting in place (case 1) and superior moving towards the exit (case 2). The results show that the counter flow will always reduce the evacuation efficiency to a certain extent, and the evacuation efficiency of case 1 is lower than that of case 2. At the same time, for these two cases, increasing the number of rescuers increases the evacuation time. We also find that the existence of counter flow can enlarge the effect of “faster-is-slower”, while increasing the number of exports can significantly improve the rescue efficiency. We hope that this result can help to improve the efficiency of emergency evacuation with rescue.

This paper presents a novel approach to quantitatively analyzing pedestrian congestion in evacuation management based on the Hughes and social force models. An accurate analysis of crowds plays an important role in illustrating their dynamics. However, the majority of the existing approaches to analyzing pedestrian congestion are qualitative. Few methods focus on the quantification of the interactions between crowds and individual pedestrians. According to the proposed approach, analytic tools derived from theoretical mechanics are applied to provide a multiscale representation of such interactions. In particular, we introduce movement constraints that illustrate the macroscopic and microscopic states of crowds. Furthermore, we consider pressure propagation and changes in the position of pedestrians during the evacuation process to improve the accuracy of the analysis. The generalized force caused by the varied density of pedestrians is applied to calculate the final congestion. Numerical simulations demonstrate the validity of the proposed approach.

It is essential to reliably predict the human-induced vibrations in serviceability design of footbridges to ensure the vibration levels to be within the acceptable comfort limits. The human-induced structural responses are dependent on the dynamic properties of structures and human-induced excitations. For concrete footbridges, the elastic modulus of concrete is a vital parameter for determining the dynamic structural properties. To this end, a two-stage machine learning (ML)-based method is first proposed for modeling the elastic modulus of concrete. At the first stage, the ensemble algorithm, i.e. the gradient boosting regression tree (GBRT), is used to predict the compressive strength by selecting eight parameters, including concrete ingredients and curing time, as the inputs. At the second stage, the elastic modulus of concrete is modeled by using the GBRT method with the compressive strength as the input. Pedestrian crowd-induced load is the most common and crucial design load for footbridges. To consider the inter- and intra-subject variability in walking parameters and induced forces among persons in a crowd, a load model is developed by associating a modified social force model with a walking force model. By integrating the two submodels of structure and excitation, an intelligent analysis method for human-induced vibration is finally developed. A concrete footbridge with typical box cross-section subjected to human-induced excitation is analysed to illustrate the application of the proposed method.

Based on suitable video recordings of interactive pedestrian motion and improved tracking software, we apply an evolutionary optimization algorithm to determine optimal parameter specifications for the social force model. The calibrated model is then used for large-scale pedestrian simulations of evacuation scenarios, pilgrimage, and urban environments.