Public transit agencies often select time points based on operational experience rather than empirical evidence. This study addresses that gap by analyzing General Transit Feed Specification (GTFS) static datasets from four Canadian cities: Calgary, Vancouver, Ottawa, and Edmonton, to identify the operational and contextual factors that influence time point placement.

For each city, key route-level variables were defined: total number of stops, route length, trip duration, average headway, transfer stops (within a 300m radius), and proximity to key locations such as schools, hospitals, and major activity hubs. After validating GTFS-labeled time points against official agency documentation (95.3% match accuracy), correlation matrices were constructed per city, multicollinearity was assessed using Variance Inflation Factors (VIF), and simple and multivariate OLS regression models were developed. Model performance was evaluated using R² and RMSE.

Results revealed distinct patterns across cities. In Calgary, number of stops and route length were the strongest predictors. In Vancouver, high-frequency service (short headways) reduced the need for time points. In Ottawa, transfer stops and key locations dominated, reflecting its hub-based network structure. In Edmonton, route length alone explained most of the variation, consistent with its large urban footprint. Across all cities, transfer-related variables emerged as the most consistent predictors, confirming that network connectivity drives time point placement decisions.

These findings highlight the value of integrating open transit data into scheduling practice. Future work includes applying GTFS-RT and AVL data for real-time analysis, and expanding the framework to additional city types to assess scalability.