loading…
Buses only. Each leg is timed stop to stop, dwell included, against its own quickest hour.
Real vehicles, timed between your two ends — anywhere on a route, not just at its stops. Standing at stops and waiting at lights is included; waiting for the bus to turn up is not. Averages add up to a journey and medians do not, so a long ride reads about a tenth quicker on the median. Drag either end to move it, Esc to clear.
How long Lviv city buses take to get from one stop to the next, and how much longer that takes at a given hour than at the quickest hour of the day. Every stop-to-stop leg of the network is drawn along the street the buses drive, nudged to the right of its direction of travel so the two directions of a street lie side by side. Green means the leg runs about as quick as it ever does; red means it takes twice as long or more. Use the hour slider to compare the morning peak against the evening, the month picker to compare months, and the day selector to separate weekdays from weekends.
Each leg is compared only with itself: its off-peak time is its quickest hour of the day with at least ten timed rides. That separates congestion from distance — a long leg on the edge of town is not red for being long, and a short one in the centre is red only if it takes much longer at rush hour than it does at its quietest. Where several routes share a leg, their times are pooled, weighted by how many rides each contributed. Clicking a leg shows its ride time, its off-peak time and hour, every route that drives it, and its ride time hour by hour. The Statistic selector opens on the median, which is how a leg usually goes; the average is there too.
Drop a point where you are and another where you are going, and the map lists every bus route that drives past both — in that order, with an arrow showing which way — timed on what its vehicles actually did between those two points. Neither end has to be a stop: each point is placed on the route's own line, so a spot halfway down a block is timed as a spot halfway down a block. The figures come from the same pass over the archive that follows individual buses between consecutive stops, so they include the dwell at each stop and the wait at each light. Picking a route draws the stretch of it being timed. Waiting for the bus to turn up is not measured: the archive records where buses were, not when the next one leaves.
Only buses are counted. Lviv trolleybuses are published in the schedule feed as
route_type=3, the same code as buses, so they are excluded by route name;
trams are excluded by route type. A leg is timed from the moment a vehicle comes
within the stop radius of one stop on its own trip to the moment it reaches the next,
so standing at the stop and waiting at lights are part of the leg. A leg seen on fewer
than five rides in a slot is left out rather than guessed.
Source data: the Lviv public transport GTFS-RT feed, archived to object storage. Basemap: OpenStreetMap.