Although automatically collected human travel records can accurately capture the time and location of humanmovements, they do not directly explain the hidden semantic structures behind the data, e.g., activity types. This work proposes a probabilistic topic model, adapted from Latent Dirichlet Allocation (LDA), to discover representative and interpretable activity categorization from individual-level spatiotemporal data in an unsupervised manner. Specifically, the activity-travel episodes of an individual user are treated as words in a document, and each topic is a distribution
over space and time that corresponds to certain type of activity. The model accounts for a mixture of discrete and continuous attributes---the location, start time of day, start day of week, and duration of each activity episode. The proposed methodology is demonstrated using pseudonymized transit smart card data from London, U.K. The results show that the model can successfully distinguish the three most basic types of activities---home, work, and other. As the specified number of activity categories increases, more specific subpatterns for home and work emerge, and both the goodness of fit and predictive performance for travel behavior improve. This work makes it possible to enrich human mobility data with representative and interpretable activity patterns without relying on predefined activity categories or heuristic rules.

High-frequency transit systems are essential for the socioeconomic and environmental well-being of large and dense cities. The planning and control of their operations are important determinants of service quality. Although headway and optimization-based control strategies generally outperform schedule-adherence strategies, high-frequency operations are mostly planned with schedules, in part because operators must observe resource constraints (neglected by most control strategies) while planning and delivering service. This research develops a schedule-free paradigm for high-frequency transit operations, in which trip sequences and departure times are optimized in real-time, employing stop-skipping strategies and utilizing real-time information to maximize service quality while satisfying operator resource constraints. Following a discussion of possible methodological approaches, a simple methodology is applied to operate a simulated transit service without schedules. Results demonstrate the feasibility of the new paradigm.

The near ubiquity of smartphones has the potential to transform how researchers, companies, and public transit agencies understand travel behavior. This research analyzes how an emerging class of automatically-collected data based on smartphone GPS and sensor information - referred to here as mobile activity-tracking data - can be used in a transit agency to better understand travel behavior. Through a collaboration with Transport for London, multiple weeks of mobile activity-tracking data of London residents was collected between 2015 and 2016 using an application called Moves. Using this case study, this paper discusses the benefits of this new data and how it compares with other data at TfL and elsewhere and examines the process of collecting the data. Using the resulting data, this paper then compares the resulting trip records from the mobile activity tracking data with those from the automatic fare card data collected during the same period and same individuals. By comparing mobile activity tracking with an established, well-researched data source like AFC, the authors observe that while the trip match rate between the two data sources is high (68%) but not perfect. Next, the paper proposes a probabilistic framework to identify between motorized trip modes using mobile activity tracking data and and the public transit network. Specifically, the model uses both spatial characteristics, such as distance to public transit network, and trip characteristics such as speed in order to identify the trip mode as bus, rail, subway, or motorized non-public transit. Using logistic regression, classification tree, and random forest, this model achieves an accuracy of 90%, 91%, and 92% respectively.