When high-quality information provided by a transit agency isn’t available, Google uses aggregated crowdsourced data to predict public transit:
- Real-time locations
- Arrival and departure times
- Crowd size
Live trip status for trains
Location history reports from users around the world are combined with train track locations to detect moving trains. The trains are matched to published schedules to predict when the trains will arrive at upcoming stations. This data is provided to Google Maps users in the form of real-time train locations and arrival and departure times.
Live traffic for buses
Google forecasts bus travel times by using a machine learning model that combines real-time car traffic forecasts with data about bus routes and stops.
It’s not enough to simply calculate car traffic because buses navigate in a fundamentally different way than cars. For example, buses may:
- Stop at bus stops
- Take longer to accelerate, slow down, and turn
- Have special road privileges like dedicated bus lanes
This difference in movement is factored into the model to create accurate delay information tailored specifically for buses. Get more technical details on our Google AI blog.