Work

Modelling evacuation mode choice behaviour under an imminent threat of volcanic eruption using logistic regression and machine learning methods

Year: 2026

Type: article

Abstract: With the increasing frequency of natural hazards worldwide, developing effective evacuation strategies is crucial to reducing their adverse impacts. This study examines the determinants of evacuation mode choice in the event of an imminent volcanic eruption in Auckland, New Zealand, using data from a stated preference survey. Key influencing factors, including socio-demographics, risk perception, evacuation timing, and destination choice, are analysed using logistic regression and machine learning models (Random Forest, Support Vector Machine, Probabilistic Neural Network, and Classification and Regression Tree). The results identify vehicle ownership, evacuation timing, and route/destination preferences as critical determinants. Among the models, Random Forest demonstrates the highest predictive accuracy, outperforming logistic regression. The findings provide valuable insights for emergency planners and policymakers to design more resilient and adaptive evacuation strategies.

Source: Transportation

DOI: 10.1007/s11116-026-10818-7

Authors: Snehalata Thakur, Prakash Ranjitkar, Soroush Rashidi

Institution: University of Auckland

Language: English

Topic: Evacuation and Crowd Dynamics

Subfield: Ocean Engineering

Field: Engineering

Domain: Physical Sciences

SDGs: Sustainable cities and communities, Good health and well-being

Open Access status: hybrid

Estimated APC paid: $2,990