This paper discusses the challenges and opportunities of using large, automatically collected data sources for the purpose of measuring and monitoring transit system performance over time, demonstrating factors such as changes in reliability and ridership.

It discusses work undertaken within the initial stages of ongoing research for the analysis of transport in San Francisco using the city’s public transport vehicle location and passenger counting data, expanded with openly-available timetable information. It provides examples of open-source software tools developed for the fusion, analysis and visualisation of these large data sources. Such tools enable explorative and retrospective analyses of the city’s transport network and provide opportunity of further work and research into network performance and integration of multiple other large data sources.

The paper investigates and highlights methods whereby data can be manipulated in order to assist in the creation of long-term data inventories of travel behavior and transport performance. These methods can be applicable to the data landscape for planning and modelling of transport systems in Australian cities.