Testing the Hurricane Evacuation Modeling Package (HEMP)

Bian, Ruijie “Rebecca” · 2025 · 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 evaluates the Hurricane Evacuation Modeling Package (HEMP), a tool developed by the Louisiana Transportation Research Center (LTRC) to predict evacuation patterns and traffic impacts. The primary research question addresses whether behavioral models and traffic simulation frameworks, originally estimated from past storm data, can be effectively transferred to predict evacuation behavior in new storm settings, specifically Hurricane Ida (2021). The motivation is to enhance the tool’s accuracy and computational speed for potential real-time application in emergency management and digital twin environments. The methodology involved two main components: simulating evacuation-related choices and simulating evacuation traffic. For behavioral simulation, the researchers generated synthetic populations for the New Orleans metropolitan area for years 2013–2022 using publicly available data. They applied pre-estimated behavioral models—including evacuate/stay, departure timing, mode, accommodation, and destination choices—to the 2021 synthetic population to simulate household decisions during Hurricane Ida. These simulated outcomes were compared against post-storm survey data to assess model transferability. For traffic simulation, the study tested 20 different scenarios using a traffic assignment model with embedded driver route choice parameters. The goal was to identify the model configuration that best matched observed traffic volumes from loop detectors during the actual evacuation. Key findings indicate significant limitations in model transferability without specific updates. The study found that evacuation behavior simulations are only viable for storms occurring between 2013 and two years prior to the current year, as population migration renders older demographic data inappropriate for current simulations. To improve prediction accuracy, the researchers identified two critical factors requiring updates: the lognormal distance function parameters in the evacuate/stay and departure timing models, and the destination risk perception values in the destination choice model. Both updates are tied to storm characteristics and require live storm feeds, highlighting the indispensability of real-time data input. In the traffic simulation phase, the stochastic shortest path model, which minimizes travel time and assumes 50% of drivers are informed, provided the best fit for observed traffic patterns during Hurricane Ida. The significance of this research lies in its demonstration of the necessity for integrating real-time data—such as population profiles, storm forecasts, and near-real-time background traffic—into evacuation simulations. The study concludes that while pre-estimated statistical models provide a foundation, their practical utility depends on updating specific parameters related to storm dynamics and current demographics. This work emphasizes the importance of incorporating human components, including demographic shifts and choice behaviors, into digital twins to support more accurate and effective disaster management strategies.

Key finding

Updating lognormal distance function parameters and destination risk perception values with real-time storm data, combined with a stochastic shortest path traffic model, significantly improves the accuracy of hurricane evacuation simulations.

Methodology

modeling

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.

• mental model of traffic