Pre-evacuation Time Estimation Based Emergency Evacuation Simulation in Urban Residential Communities

Area of Science:

• Urban planning and disaster management
• Civil engineering and safety science
• Human behavior in emergencies

Background:

• Community-scale emergency evacuation planning requires accurate estimation of evacuation time components.
• Pre-evacuation time, a critical factor in residential community evacuations, is often overlooked in existing studies.
• Realistic evacuation simulations are essential for ensuring resident safety during emergencies.

Purpose of the Study:

• To propose a novel method for estimating pre-evacuation time in urban residential communities.
• To investigate factors influencing pre-evacuation time using survey data and develop a predictive model.
• To enhance the realism and accuracy of community-scale emergency evacuation simulations.

Main Methods:

• Integration of resident pre-evacuation behavior data from surveys.
• Development of a predictive model for pre-evacuation times using the Random Forest algorithm.
• Conducting sensitivity analysis to identify critical parameters for evacuation simulations.

Main Results:

• The proposed method significantly enhances the realism of urban residential community evacuation simulations.
• Pre-evacuation time was found to have a substantial impact on total evacuation time, congestion time, and congestion degree.
• A case study in Shanghai demonstrated the method's feasibility and effectiveness.

Conclusions:

• Accurate estimation of pre-evacuation time is vital for effective urban community emergency management.
• The developed model provides a valuable tool for predicting pre-evacuation times and improving evacuation strategies.
• This research contributes to a deeper understanding of resident behavior during emergencies and enhances overall community safety planning.