Using GPS data to explore speed patterns and temporal fluctuations in urban logistics: The case of São Paulo, Brazil

Large-scale urban sensing data, such as vehicle global positioning system (GPS) traces, are emerging as an important data source for urban planning. However, the widespread availability of commercial truck GPS data comes with a large diversity of sources with different operational characteristics. In this context, motivated by addressing ongoing challenges in urban logistics from the perspective of transport geography, we describe an approach to exploring and analyzing a large amount of vehicle GPS data from heterogeneous sources. Our aim is to leverage GPS data to better understand and characterize urban logistics from a broader, city-wide perspective as well as from the perspective of a single corporate logistics system. Using the city of São Paulo, Brazil, as an example, we show how such data can be leveraged to improve the operational planning of transport operators and the policymaking strategies of public agencies, especially in cities in developing and emerging economies where other sources of data are not readily available. We build on GPS data from company-specific logistics fleets as well as company-agnostic, mixed-use general vehicle fleets to present novel insights into the spatial distribution of cargo vehicles across the city on different days, the distinct flows and preferred paths of freight traffic, the number of stops per vehicle trip, the distribution of stop times, and the speed patterns. Our proposed approach can easily be transferred and extended to other geographic contexts and other datasets that exhibit structures similar to those used in our case study.