| Resumo: | This paper explores the potential of using crowdsourcing tools, namely Google "Popular times" (GPT) as an alternative source of information to predict traffic-related impacts. Using linear regression models, we examined the relationships between GPT and traffic volumes, travel times, pollutant emissions and noise of different areas in different periods. Different data sets were collected: i) crowdsourcing information from Google Maps; ii) traffic dynamics with the use of a probe car equipped with a Global Navigation Satellite System data logger; and iii) traffic volumes. The emissions estimation was based on the Vehicle Specific Power methodology, while noise estimations were conducted with the use of “The Common Noise Assessment Methods in Europe” (CNOSSOS-EU) model. This study shows encouraging results, as it was possible to establish clear relationships between GPT and traffic and environmental performance.
|
|---|