Profiling urban activity hubs using transit smart card data

Understanding why and where people travel by public transport is a key enabler for smart cities because it informs city planning, daily operations, and sustainable city growth. This article introduces a data-driven approach using transit smart card data to discover where activities are concentrated and why people travel to those regions. Our approach is based on the idea of stays between passenger trips. A stay has an arrival time and a period of time spent in a certain region. The regions where stays are concentrated are called hubs. Coherent clusters of stays indicate human activities such as going to work or short errands. An efficient and robust algorithm is proposed for learning hubs and their activities. Triangulation with points of interest and ticket data validates that activity stays and hub activities satisfy common sense expectations. The utility of the activity hub profiles for urban planners and transport managers is demonstrated by use cases for operational, tactical and strategic goals.