Smart Connected and Automated Vehicle Fleet Management: Developing Regional Dispatch Decision Support for Congestion Mitigation
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 research addresses the increasing workload and operational complexity faced by regional transportation dispatchers due to the emergence of connected and autonomous vehicles (AVs). As AVs integrate into mixed traffic, dispatchers must manage not only traditional congestion but also potential safety hazards, such as vehicle breakdowns that block emergency responders. The study aims to develop decision support tools to assist dispatchers in strategic (weeks to months) and tactical (hours to days) planning, while simultaneously evaluating the safety implications of AV platooning and the accuracy of simulation models used to predict AV behavior. The study employed a three-pronged approach. First, the researchers developed the Congestion Alerting Decision Support (CADS) tool using the open-source microsimulation software SUMO, selected for its flexibility and ability to represent behavioral uncertainty compared to commercial alternatives like PTV VISSIM. CADS allows users to run "what-if" simulations to assess congestion impacts from events like weather or construction and to determine optimal resource allocation. Second, the team conducted field tests involving mixed platoons of traditional and autonomous vehicles to analyze traffic conflicts using Surrogate Safety Measures (SSMs). Third, the researchers evaluated three widely used car-following models—Adaptive Cruise Control (ACC), Intelligent Driver Model (IDM), and Wiedemann 99 (W99)—by comparing their predictions against real-world AV trajectory data. Key findings indicate that while CADS effectively visualizes congestion and aids in planning, it currently lacks the capability to link congestion metrics to safety predictions. Field tests revealed that AVs operating as following vehicles in mixed platoons exhibited longer response times, with system instability increasing as platoon length grew, potentially raising conflict severity. Regarding simulation accuracy, the study found that while all three car-following models accurately predicted AV positions over time, they struggled to replicate real-world acceleration and deceleration profiles. The ACC model most closely matched real-world acceleration, whereas W99 produced abrupt, unrealistic speed changes, and IDM showed low acceleration noise due to conservative driving algorithms. Consequently, current models are insufficient for reliable risk projection regarding collisions. The significance of this work lies in its identification of critical gaps in current dispatcher support tools and AV simulation methodologies. The results suggest that while existing simulations can predict vehicle location, they require further refinement to accurately model dynamic driving behaviors for safety analysis. The authors recommend future work focus on operationalizing CADS, incorporating non-linear car-following behaviors, and improving model calibration to better predict high-risk areas. This research provides a foundation for enhancing regional dispatch capabilities, ensuring that transportation authorities can effectively manage the safety and congestion challenges posed by the growing integration of autonomous fleets.
Key finding
Autonomous vehicles operating as followers in mixed-traffic platoons demonstrate longer response times and increased instability compared to traditional vehicles, and while standard car-following models accurately predict vehicle positions, they fail to accurately simulate acceleration and deceleration behaviors needed for reliable safety risk projections.
Methodology
mixed_methods
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).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 24 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
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.
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