Integrating supply and demand aspects of transportation for mass evacuation under disasters.

Peeta, Srinivas; Hsu, Yu-Ting · 2009 · ROSA P / NEXTRANS Center (U.S.)

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 mass evacuation modeling regarding the integration of supply-side infrastructure management and demand-side evacuee behavior, specifically within the context of no-notice disasters. While previous research predominantly focused on supply-side solutions like contra-flow lanes or relied on simplified, static demand assumptions, this work argues that effective evacuation strategies require a dynamic understanding of how individuals respond to real-time conditions. The research is motivated by the operational complexity of no-notice events—such as earthquakes or terrorist attacks—where evacuees face extreme time pressure, uncertainty, and existential threats, leading to behavioral patterns distinct from those observed in short-notice scenarios like hurricanes. To model these behaviors, the authors developed a comprehensive framework focusing on aggregate behavior at the Traffic Analysis Zone (TAZ) level, acknowledging practical data limitations during real-time operations. The methodology employs discrete choice theory, specifically incorporating fuzzy set theory into mixed logit models to account for subjective risk perceptions, linguistic information descriptions, and individual heterogeneity. The model is hierarchical, addressing two primary decisions: first, the evacuation participation decision, which determines whether an individual evacuates at a given time based on perceived risk, emergency management agency (EMA) recommendations, herding behavior, and state dependence; and second, the route choice decision, which selects paths to the nearest safe places based on estimated travel time, perceived route risk, EMA guidance, and a documented "freeway bias." The study validated these models through simulation experiments designed to test their prediction capabilities and robustness. Numerical experiments included sensitivity analyses regarding demand levels and background traffic ratios. The results demonstrated that the proposed fuzzy mixed logit models could effectively interpret evacuation behavior from observable variables at an aggregate level, capturing the interplay between demand and supply dynamics. The simulations confirmed the models' ability to handle the randomness and fuzziness inherent in disaster dynamics and evacuee decision-making processes under time pressure. The significance of this work lies in its provision of a behavior-robust platform for designing information strategies for emergency management agencies. By formally linking behavioral insights with operational control, the study moves beyond the assumption of full driver compliance, offering a more realistic basis for dynamic routing and congestion mitigation. The authors conclude that while aggregate models are necessary for immediate deployment, future research should focus on disaggregate field surveys to refine individual-level behavior models. Ultimately, this framework supports the development of behavior-based control models that can optimize evacuation clearance times and enhance system efficiency during catastrophic events.

Key finding

Simulation experiments demonstrate that the proposed fuzzy mixed logit models can effectively interpret and predict aggregate evacuation participation and route choice behavior from observable variables under no-notice disaster conditions.

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.

• driver post crash behavior