Pedro started his journey in data science and urban analytics in 2016 through the Cloud Computing for Big Data Centre for Doctoral Training (CDT) PhD programme at Newcastle University. After one year of intensive courses, he was introduced to the challenges faced by traffic operators and transportation planners working towards eradicating traffic congestion in urban environments. Motivated by exciting advances in the world of transportation and urban mobility, he pursued the opportunity to work in the complex and interdisciplinary real-world problem of traffic monitoring and prediction. Fuelled by new collaborations within a growing research community, he leverages a newer breed of traffic cameras and state-of-the-art statistical methods to develop a better understanding of how traffic congestion behaves within our cities.

Prior to starting his PhD, Pedro worked in a variety of research problems ranging from healthcare and wearable sensors, resulting in the Master thesis “Towards a real-time pervasive system for blood pressure monitoring”, to fog and edge computing as a a research assistant at University of Porto.

Research interests

In the last decade, there has been an explosion in the amount of transportation data available for studying human mobility. Companies such as Google, Waze and TomTom employ real-time data-centric approaches that have revolutionised how people plan and execute their everyday journeys. This paradigm shift is also occurring at the level of government and local authorities, responsible for developing short and long-term strategies for tackling traffic congestion. The growing number of urban sensors means that we can better monitor, study and understand everyday urban processes and their impact on quality of life. Traffic cameras, in particular, produce large amounts of exhaust data that can be used to develop a deeper understanding of traffic patterns in cities and urban environments. Is traffic congestion contagious? How fast and far does congestion propagate? Where does congestion repeatedly occur and what role does network topology and demand play? Questions such as these are not addressed by Google Traffic, but can have a big impact on how traffic professionals design new roads and program traffic light timers in reaction to high demand and incidents.

Coming from a different background, Pedro aspires to expand his knowledge of urban processes and urban theory. To draw inferences from data, he aims to develop his knowledge of Bayesian Statistics and his ability to apply it to mobility and spatio-temporal data. With a strong passion for reproducibility, Pedro works towards packaging and releasing software that can be used not only to engineer and transform raw data, but also to explore and model traffic flow data. More importantly, he is actively looking to foster new relationships and collaborations in these areas so that his work has real outputs and benefits.