Study on Pedestrian Movement and Emergency Evacuation under the Effect of Leader Behavior

Abstract

In recent years, with the continuous development of urbanization, the number of buildings with complex structures has increased. This situation dramatically challenges daily crowd management and safety evacuation in emergencies. Therefore, it is essential to have well-trained leaders familiar with the layout and routes in emergency evacuations. Studying pedestrians’ movement characteristics and evacuation behavior under the effect of leader behavior is crucial to improving crowd management and evacuation efficiency. However, few experimental studies exist on the characteristics of evacuation movement of pedestrians under guidance. Moreover, the effect of static guidance on pedestrian evacuation has not been adequately addressed in model studies.

This thesis focused on pedestrian movement and evacuation through a series of controlled experiments. The research comprised two primary aspects. Firstly, it addressed the influence of movement changes of the leader on crowd movement characteristics, revealing the mechanism of movement coordination between the leader and the followers. Secondly, it explored the characteristics of pedestrian evacuation under static guidance, quantifying the effects of stress level and guidance positions on evacuation efficiency. Based on the controlled experiment findings, a social force model was utilized to propose an evacuation model with guidance. This model was then applied to analyze the influence of guidance arrangements on evacuation efficiency in subway station halls during emergencies. To summarize, this thesis's main objectives and results are as follows.

Firstly, experiments were conducted to examine the behavior of crowds following a leader in a two-dimensional setting. The aim was to understand how changes in the leader's movement would affect the movement characteristics of the following crowd. The findings revealed that the crowd exhibited greater instability as the leader's speed increased. This condition was evident through differences in speed fluctuations, group structure, and the evolution of walking radius and local density. Moreover, the numerical relation between the relative motion parameters of the leader and followers was established to reveal the mechanism of movement coordination between them. Subsequently, a method was proposed to evaluate the leadership effect in walking groups. Weight parameters were defined to evaluate the level at which followers are in speed matching, angular speed adjustment, and ineffective leadership states. Normalization parameters were used to evaluate the overall coordination of the crowd. The results indicated that increased group size and leader speed significantly weakened the leadership effect. Furthermore, crowd coordination was found to be more affected by group size than by leader speed.

Secondly, experiments were conducted to examine pedestrian evacuation dynamics under static guidance and the evacuation characteristics of pedestrians, explicitly focusing on the influence of leader behavior. The pedestrian evacuation efficiency and the difference in movement characteristics such as speed, density, and flow rate under different stress levels and guidance positions were quantified. The results indicated that guidance under high incentive levels positively affected pedestrian evacuation. Moreover, the overall evacuation efficiency was higher when the pedestrians who could benefit from guidance were located further away from the correct exit. Conversely, the effect of guidance under low-stress levels on pedestrian evacuation was not crucial. Additionally, the choice of exit by pedestrians was influenced by physical and social factors. Pedestrians with high-stress levels took longer to make decisions s than those with low-stress levels. These findings may highlight the intricate relationship between stress, guidance, and decision-making during pedestrian evacuation.

Based on the insights gained from the experiments on pedestrian evacuation characteristics under static guidance, an evacuation model with static guidance was proposed, addressing the leader’s influence and the pedestrians’ decision-making behavior. Then, the model was validated using experimental data, indicating that the model could simulate well the evacuation process of pedestrians in rooms with obstacles and unusable exits. Subsequently, the model was employed to investigate the effect of different locations and numbers of static leaders on evacuation efficiency. It was found that a leader in the center of each potential exit exhibited a better leadership effect in the scenario with unusable exits, and the farther the guided pedestrian was from the correct exit, the better the pedestrians’ evacuation performance. The distance parameter of multiple leaders was defined, and the optimal number of leaders exited in a specific scenario due to the overlap of leaders’ influencing areas. Based on the questionnaire and field survey data, the model was implemented to simulate the evacuation movement of pedestrians in the subway station hall, suggesting leaders' positive impact on emergency evacuations. Various factors such as view distances, guidance numbers, guidance locations, and the proportion of pedestrians accepting the guiding information, were explored to assess their effects on evacuation efficiency.

The findings from the experiments on pedestrian evacuation movement under guidance offer valuable data support for informing safety research on crowd evacuation guidance. These insights can provide a scientific basis for developing crowd management schemes in public places. Additionally, the proposed evacuation guidance model offers scientific support for devising effective strategies for crowd evacuation, including the optimal arrangement of leaders in evacuation scenarios.

Date of Award	10 Jul 2023
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Lizhong Yang (External Supervisor) & Siu Ming LO (Supervisor)