System Dynamics Models of Automated Vehicle Impacts

Smith, Scott; Rakoff, Hannah; Eilbert, Andrew; Shaw, Jingsi; Berg, Ian · 2023 · ROSA P / United States. Department of Transportation. Intelligent Transportation Systems Joint Program Office

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 report addresses the challenge of modeling the transformative and uncertain impacts of automated driving systems (ADS) on transportation systems. Traditional travel demand models are often too rigid and slow to explore a wide variety of potential futures involving automation, electric vehicles, and telework. To address this, the study employs System Dynamics (SD), a methodology capable of capturing complex feedback effects and changes over time. The research aims to provide both qualitative frameworks for stakeholder alignment and quantitative tools for scenario planning, specifically focusing on the viability and impacts of shared automated mobility services. The methodology combines qualitative and quantitative SD approaches. Qualitatively, the researchers used Group Model Building (GMB) with diverse stakeholders, including U.S. and European agencies, to develop Causal Loop Diagrams (CLDs). These diagrams identified key system archetypes, such as technology adoption, business model sustainability, mode choice, scale effects, congestion, and land use. Quantitatively, the team constructed a stock-and-flow model of a shared mobility service, integrating perspectives from both service operators (financial sustainability) and travelers (service attractiveness). The model was calibrated using Transportation Network Company (TNC) data from Massachusetts and Chicago. It was then tested across urban, suburban, and rural scenarios, with sensitivity analyses examining variables such as trip density, value of time (VOT), and induced travel. The findings indicate that higher trip densities in urban areas lead to lower wait times and greater returns on investment, making shared automated services more attractive to both users and operators in these regions. The model demonstrated that Vehicle Miles Traveled (VMT) increases primarily due to induced travel, a factor significantly influenced by the value of time. In rural areas, high-quality service with low wait times is possible, except at the lowest population densities. The qualitative CLDs successfully mapped the causal mechanisms driving these outcomes, such as the reinforcing loop where scale improves service efficiency, and the balancing loop where congestion reduces attractiveness. The study also confirmed that cooperative adaptive cruise control can smooth driving and reduce emissions, though this requires specific controller settings. The significance of this work lies in providing a flexible, fast-running modeling framework that bridges the gap between planners and modelers. By using SD, the study offers a way to test policy levers and understand complex feedback loops that traditional models may miss. The report concludes by suggesting future work on vehicle ownership dynamics and the integration of SD techniques with existing strategic models like VisionEval and POLARIS. This approach supports performance-based planning by allowing agencies to explore uncertain futures and identify potential tipping points in transportation system behavior.

Key finding

Higher trip densities in urban areas result in lower wait times and greater returns on investment for shared automated mobility services compared to suburban and rural areas.

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.

• acceptance adoption
• driverless ads
• passenger motion sickness comfort
• exposure measurement

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Theoretical Contribution: computational model