Modeling Hurricane Evacuation Traffic: Development of a Time-Dependent Hurricane Evacuation Demand Model [Report]

Fu, Haoqiang; Wilmot, Chester · 2006 · ROSA P / Louisiana Transportation Research Center

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study addresses the critical gap in transportation planning regarding the estimation of dynamic travel demand during hurricane evacuations. Traditional static models fail to capture the time-dependent nature of evacuation traffic, where factors such as storm proximity and policy decisions change rapidly. The authors argue that accurate dynamic modeling is essential for effective evacuation planning, as it allows for realistic predictions of congestion and network utilization. The research aims to develop and compare dynamic travel demand models that can reproduce evacuation behavior more accurately than conventional methods using fixed participation rates and response curves. The researchers developed dynamic demand models using two primary methodologies: survival analysis and sequential choice models. Survival analysis treated the time before evacuation as a duration, modeling households that did not evacuate as censored observations using Cox proportional hazards and Piecewise Exponential models. The sequential choice approach modeled the evacuation decision as a series of binary choices over time, utilizing sequential logit and sequential complementary log-log models. These models predicted the probability of a household evacuating in each time period based on socio-economic characteristics, hurricane attributes (e.g., distance to storm), and policy variables (e.g., evacuation orders). The study utilized three datasets: revealed preference data from southwest Louisiana following Hurricane Andrew, revealed preference data from South Carolina following Hurricane Floyd, and stated preference survey data from the New Orleans area. The analysis identified the sequential logit model as the superior alternative for modeling dynamic hurricane evacuation demand. It produced predictions that outperformed current evacuation participation rate models with response curves. The study demonstrated the model's transferability by applying parameters estimated from the Hurricane Floyd data to the Hurricane Andrew dataset, achieving reasonable accuracy in a different environmental context. This suggests the model can estimate dynamic demand across varying locations and storm conditions. However, the authors noted that further research is needed to refine procedures for updating transferred model parameters to better reflect local evacuation behaviors. The significance of this work lies in its contribution to the advancement of Dynamic Traffic Assignment (DTA) for emergency management. By providing a robust method for estimating time-dependent origin-destination demand, the sequential logit model supports more accurate and realistic simulations of traffic conditions during evacuations. This capability enables emergency officials to develop, test, and compare evacuation plans and management strategies with greater precision, potentially optimizing network usage and reducing congestion. The study concludes that while the sequential logit model is effective, continued work on model transferability and local parameter calibration is necessary for broader application.

Key finding

The sequential logit model produced predictions superior to current evacuation participation rate models with response curves and demonstrated reasonable accuracy when transferred to a different geographic environment.

Methodology

dataset

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• exposure measurement