Guide-evacuee-obstacle interaction enabled evacuation dynamics from school facilities : analysis and simulation

• Main Author: Wang, Ke
• Other Authors: Qian Shunzhi
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Civil engineering::Transportation
• Online Access: https://hdl.handle.net/10356/146644
• Institution: Nanyang Technological University

Emergency evacuation is a challenging issue due to scarcity of real-case data, complexity of human behaviours, and context-dependency nature. This study is thus especially focused on the behaviour analysis and the modelling of evacuation process in the school context. Two folds of works have been conducted: 1) proposing a cost-effective analytical framework to reuse existing data with machine learning methods like decision tree, support vector machine, K-nearest neighbour and artificial neural network. Applied to stepwise movement pattern analysis, this framework is found to be robust and handy. 2) establishing a holistic evacuation model considering “guide-evacuee-obstacle” interaction in school facilities. The developed model incorporates strategic (evacuation initialization), tactical (route choice), and operational level (stepwise movement) behaviours, which serves as a tool to find behaviour optimisation and layout design strategies at each phase of the evacuation. Findings from this study would hopefully benefit academic research and practical applications in emergency evacuations.