Exploring the Potential of Web Based Information of Business Popularity for Supporting Sustainable Traffic Management

Bandeira, Jorge M.; Tafidis, Pavlos; Macedo, Eloísa; Teixeira, João; Bahmankhah, Behnam; Guarnaccia, Cláudio; Coelho, Margarida C. · 2020 · Crossref

DOI: 10.2478/ttj-2020-0004

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 potential of Google Maps’ “Popular Times” (GPT) feature as a crowdsourced data source for predicting traffic volumes and associated environmental impacts, specifically pollutant emissions and noise. Motivated by the high costs and limitations of traditional traffic monitoring systems, particularly in medium-sized cities with limited resources, the research aims to determine if web-based business popularity data can reliably estimate traffic demand and its externalities. The authors address a gap in existing literature, which has largely focused on other social media or Google traffic speed data, by testing GPT’s utility in forecasting congestion and pollution levels. The methodology involved empirical data collection in two medium-sized European cities: Aveiro, Portugal, and Badajoz, Spain. The researchers selected three commercial hotspots, analyzing eight specific road links surrounding these areas. Data collection included capturing GPT values every 15 minutes, which were converted into discrete variables on a scale of 0 to 1. Traffic dynamics were recorded using a probe car equipped with a Global Navigation Satellite System (GNSS) logger to measure speed, acceleration, and travel time. Traffic volumes were obtained via cameras and manual counts. Emissions of CO2 and NOx were estimated using the Vehicle Specific Power (VSP) methodology, which accounts for driving modes and fleet composition. Noise levels were simulated using the CNOSSOS-EU model, as field measurements were compromised by weather conditions. Linear regression models were employed to analyze the relationships between GPT values and the observed or estimated traffic and environmental variables. The results demonstrate strong correlations between GPT and traffic performance metrics. GPT showed a high correlation with parking lot occupancy (R² = 0.89), validating its ability to predict local demand. For traffic volumes, GPT explained up to 90% of the variability, with the strongest correlations observed on links directly serving shopping centers. CO2 emissions showed even higher predictability, with GPT explaining up to 98% of variability in certain links. Noise levels were also well-predicted, with GPT accounting for up to 85% of variability. However, the correlation for NOx emissions was weaker (up to 76%), attributed to the high sensitivity of NOx to individual driving behaviors such as acceleration and deceleration. The study also noted seasonal variations, indicating that regression coefficients may differ between winter and summer, suggesting that GPT’s predictive power is context-dependent. The findings confirm that GPT can serve as a reliable, cost-effective alternative for traffic management and environmental monitoring, particularly in areas where traditional infrastructure is lacking. The authors conclude that while GPT is effective for estimating traffic volumes, CO2, and noise, its application requires careful consideration of seasonal and locational factors. This approach offers significant implications for sustainable transport management, enabling authorities to anticipate congestion, optimize intelligent transport systems, and integrate real-time environmental data into eco-routing platforms. Future work aims to develop global models for broader street network applications.

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