Using Advanced Analytics to Understand Traffic Flow

BADEA, Gheorghe; BADEA, Ana Cornelia · 2025 · Crossref

DOI: 10.29302/revcad.2025.38.2

archive: archived pipeline: cataloged verified

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

Summary

This study investigates the application of advanced geospatial analytics to understand traffic flow patterns and their impact on emergency service accessibility in Bucharest, Romania. Motivated by Bucharest’s high congestion levels—ranked fifth globally with a 48% congestion index and resulting in 150 hours of annual time lost per driver—the research aims to correlate temporal traffic variations with queue lengths at critical intersections. The authors also seek to evaluate the capabilities of ArcGIS ModelBuilder Beta for automating such analyses. The methodology utilizes TomTom traffic services and ArcGIS Online (AGOL) to analyze floating car data (FCD) derived from GPS devices. The researchers employed junction analysis, route monitoring, and drive time area tools to assess traffic conditions. Specifically, they examined two major intersections in Bucharest’s 2nd District: the junction of the A3 highway with Petricani and Fabrica de Glucoza streets, and the intersection of Lacul Tei Boulevard, Petricani Street, and Doamna Ghica Street. The analysis covered the period of March 17–23, 2025, using Data Pipelines and ModelBuilder Beta to automate workflows. Traffic speeds were categorized by percentage of free-flow speeds, and historical profiles were compared against real-time data. Key findings reveal significant temporal and spatial variations in congestion. At the Lacul Tei Boulevard–Petricani–Doamna Ghica intersection, queue lengths reached a maximum of 970 meters on Saturdays at noon, with a transit volume of 3,582 vehicles per hour. Traffic jams exceeding 2 kilometers were observed on Lacul Tei Boulevard during rush hours on Mondays and Fridays, while queues of up to 4 kilometers occurred on Petricani Street mid-week. The study further analyzed emergency response capabilities, determining that fire stations in the 2nd District could not secure a large portion of the district within five minutes during peak congestion on Wednesday evenings. However, expanding the response window to ten minutes allowed for coverage of the entire sector. Comparisons between Sunday afternoon scenarios showed distinct differences in reachable areas for 10-minute versus 15-minute response times. The authors conclude that while current analytics cannot simulate emergency vehicle privileges (e.g., sirens and lights), the data demonstrates that severe congestion significantly restricts five-minute emergency response zones. Nevertheless, a ten-minute response time appears sufficient to cover the sector. The study highlights the utility of ModelBuilder Beta for creating repeatable, low-code workflows, despite limitations such as the inability to easily copy model inputs for parameter adjustments. These findings underscore the importance of integrating real-time traffic analytics into urban planning and emergency service logistics to improve decision-making and safety.

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.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

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

• traffic density