Traffic Analysis in a Smart City

Bachechi, Chiara; Po, Laura · 2019 · Crossref

DOI: 10.1145/3358695.3361842

archive: archived pipeline: cataloged verified

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

Summary

This paper addresses the challenge of urban mobility in smart cities by analyzing traffic flow patterns to identify trends, anomalies, and critical congestion points. Motivated by the need for efficient, sustainable urban management, the authors focus on the city of Modena, Italy. The study aims to understand daily traffic dynamics, compare weekday versus weekend behaviors, and detect significant deviations caused by special events or weather conditions. By providing a comprehensive overview of the traffic system, the research seeks to assist city planners in managing mobility and raising citizen awareness regarding sustainable transport choices. The methodology employs a micro-simulation model using the open-source SUMO (Simulation for Urban Mobility) software. The model is calibrated using real-time data collected every minute from 300 traffic sensors distributed across Modena’s road network. This input allows the simulation to generate detailed outputs for over 800 km of roads, including vehicle counts, lane density, and average speeds. To analyze the resulting spatio-temporal data series, the authors utilize Piecewise Aggregate Approximation (PAA) to reduce data dimensionality and Fast Dynamic Time Warping (FastDTW) to calculate the distance and similarity between different simulation runs. This approach enables the comparison of daily simulations against an established average weekday trend and the identification of spatial anomalies. The results reveal distinct traffic patterns based on the day of the week. Weekdays (Monday–Friday) exhibit a consistent morning peak between 7:00 and 9:30 AM, while Saturday shows a lower, earlier peak around 8:00 AM, likely due to school schedules. Sundays show no morning peak. The analysis identified specific anomalies: November 5th and 6th showed significantly lower traffic volumes and higher deviation from the average trend, which was attributed to rainy weather and a civil protection alert for river flooding. Conversely, November 1st (All Saints’ Day) showed increased traffic near cemeteries and decreased traffic on secondary roads. The study also pinpointed critical congestion areas, identifying zones with schools as the most problematic during peak hours, followed by residential and hospital areas, while industrial areas showed lower congestion. The significance of this work lies in demonstrating the effectiveness of micro-simulation models combined with time-series analysis for smart city traffic management. The findings confirm that sensor-based simulations can accurately reflect real-world traffic dynamics and detect anomalies caused by external factors like weather or holidays. By identifying specific critical points and temporal trends, the study provides actionable insights for urban planners to optimize traffic flow and infrastructure. The approach offers a scalable method for monitoring urban mobility, supporting the transition toward more efficient and sustainable city environments.

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