The methodology of the project involves 4 main steps:
1. First, each city is divided using a hexagonal spatial grid where each cell is 357 meters long (short diagonal).
2. The hexagonal grid is then used to spatially aggreagate population data from the national census, administrative records with the location of formal jobs (low-, medium- and high-qualification jobs), public schools (early childhood, primary and high school education) and publich health services (low, medium and high complexity medical care).
3. In the third step, data on public transport (GTFS format), on topography and street network are combined to calculate travel-time estimates between every pair of hexagon cells. These estimates are calculated by transport mode (walking, cycling and public transport) using OpenTripPlanner (OTP), a multimodal transportation network analysis tool. The method considers door-to-door travel-time estimates, including in-vehicle times, walking, waiting as well as transfer times.
4. The results of steps (2) and (3) are combined to calculate accessibility levels by transport mode. For the 2019 edition of the project, two accessibility indicators have been calculated:
- Minimum travel time to the closest activity
- Cumulative opportunity measure with the proportion of activities in the city that are accessible within a given time threshold (calculated for multiple thresholds).
More info about data sources and detailed methodology here.