This visualization merges all 1.4 million daily transactions using the MBTA's Charlie fare-payment card with vehicle-location data from the T's bus, subway, and light-rail fleet to infer the travel histories of that day's 0.4 million Charlie users.

Data from GPS, odometer, track circuits, and onboard bus announcements are combined to infer vehicles' stop times, and the results are merged with fare-transaction records to reconstruct each cardholder's daily travel history.

Each pixel represents a 40-meter square section of greater Boston. The brightness of each of the three RGB color components indicates the number of riders in one of three states:

Blue indicates the presence of riders prior to their first transaction of the day or after their last: it is assumed that the location of a rider's first or last transaction approximates their place of residence.

Green indicates the number of passengers in transit vehicles. On the "Vehicles" tab, buses and trains traverse the road and rail network, with their brightness indicating the number of passengers on board. The "Journeys" tab shows individual riders traveling in straight lines between the origins and destinations of their journeys--even if their journeys were inferred to have included one or more transfers.

Red indicates cardholders who are between inferred transit trips, whether transferring, engaging in activities, or traveling outside the Charlie fare system (such as by commuter rail, car, bike, or foot). Cardholders who are between inferred trips, and whose previous trip ends in a different location from where their next trip starts, are shown traveling in straight lines at constant speeds, interpolated between those two points in space and time. (On the Journeys tab, incompletely inferred transit trips are omitted from their journeys, resulting in some transit trips being shown in red as well.)

Secondary colors--cyans, magentas, yellows--result from the blending of the three cardholder states, and the comparative redness or blueness of each zone suggests different land uses.

This work is part of a research collaboration between the MIT Transit Lab and the MBTA. The visualizations will be updated as the underlying inference methods are refined and more data sources added. For more information about the algorithms and software underlying this visualization see the thesis.