Good transit: Coverage and Frequency

Posted: | Updated: | Tags: transport

When examining passenger mobility, there are many angles from which to explore effective transportation systems: service reliability, cleanliness, comfort, cost, transit coverage, service frequency, and more. While this topic is extensive, I will specifically focus on the coverage of a public transit network within an area and the significant role that frequency plays in it. I will be using the public transit system in the region of Utrecht, which is operated by Qbuzz under the name U-OV, as a demonstration. The region includes nine municipalities and a population of over 700,000. A total of 46 routes are run by U-OV, four of which are light rail and the rest are bus.

Access and Coverage

Public transit connects groups of passengers to their desired location, and scaling this to an entire city creates many potential pick-up and drop-off points.

Establishing a network of routes traveling to and from different locations maximizes access. The more stops a public transit network has, the more likely there’s a stop near where you are and where you’d like to go. Now, of course, placing too many stops would mean it takes longer for an entire trip which comes with its own problems. The goal is to place stops at locations highly frequented by passengers but also enough stops in less-frequented areas to make the transit system still viable for most trips. Placing stops closer to each other in areas of higher density, such as shopping or business districts, usually yields better capacity than placing stops at the same distance in less populated areas.

Other than the amount and location of stops, a variety of routes can be drawn up to achieve good public transit coverage over an area. A radial route connects passengers from smaller settlements to the main city center, business district, or other busy areas. A cross-city route connects two smaller settlements through a busier center; they are a combination of two radial routes. These routes not only function to move people to and from the city center but can also circulate people within the center itself on shorter journeys.

U-OV bus lines 73 and 7.

U-OV bus lines 73 and 7.

Let’s look at bus route 73, marked in red on the map, as an example. The route connects the towns of Maarssen and Zeist through the busier city of Utrecht. The route is 25 km long, with 10 stops in Maarssen, 18 in Utrecht, 1 in De Bilt, and 9 in Zeist. We can see the stops, marked by the red circles, are more concentrated in denser areas of the route, which are Maarssen, the center of Utrecht, and Zeist. The route stops at Utrecht Centraal Station which is located in the city center and a major transportation hub for busses trams within the region of Utrecht, and trains to the rest of the country and beyond.

Next let’s also take a look at a shorter bus route 7, marked in yellow on the map. Measuring 15km, the route connects the neighbourhoods of Zuilen and Westraven within the city of Utrecht. The 28 stops on this route, marked in yellow, are more evenly separated as they cross through the more densely populated city compared to route 73. In the middle of its service, it also makes a stop at the Utrecht Centraal Station.

All bus and tram lines in the Utrecht region.

All bus and tram lines in the Utrecht region.

These map lines represent all the bus routes in the Utrecht region including buses 7 and 73 talked about previously. It just comes to illustrate the coverage busses and trams have within the cities in Utrecht. Keep in mind this does not include the possible coverage and routes available through train stations like Utrecht Centraal, Utrecht Zuilen, Utrecht Overvecht, Station Maarssen, Vleuten, and others also within the region.

Frequency

Regardless of how many destinations a public transit system boasts, passengers still wouldn’t be satisfied if it meant waiting a long time to get to their destination. The more often a bus arrives at a stop the less the wait time is likely to be to the next bus. The maximum wait time between buses is called the frequency. Simply put, if a route runs at a frequency of 10 minutes, you wouldn’t need to plan your departure. If you just missed the bus, the longest you’ll have to wait is another 10 minutes. Whereas, if a bus only shows up at your stop twice an hour, you’d now have to plan your departure with the next bus making it less convenient.

The frequency of a single bus route doesn’t necessarily have to be taken into consideration in isolation. For example if you live in Zuilen, an area served by bus 7, and need to get to the city center you could use the bus to get you there. If you just missed the bus, instead of waiting another 15 minutes of the next, you could hop on bus 3, 12, or 27 depending on the next departure and get to the city center in roughly the same amount of time. So even though bus 7 only comes by every 15 minutes, you can still achieve a perceived higher frequency by utilising other routes on the network.

Interestingly, the frequency of transit is typically relative to the distance and mode of travel. For example, if you’re traveling on an urban metro system, the expectation there is for high-frequency service as riders make shorter trips throughout the city. I griped about this when writing about the up to 10 minutes you have to wait for the next metro train in Amsterdam compared to the wait time in other metro systems. On the other hand, if you’re traveling longer distances by rail from one city to the next or farther, longer wait times are usually expected.

Coverage of bus route 7 in 30 minutes starting at Prins Bernhardlaan.

Coverage of bus route 7 in 30 minutes starting at Prins Bernhardlaan.

Introducing higher frequency service also increases the opportunities to transfer to other routes, effectively extending your coverage. For example, if you lived near Prins Bernhardlaan, indicated by the red arrow, in Utrecht, and bus 7 arrived infrequently, highlighted in green are all the places you could potentially go to in 30 minutes. This is calculated based on the timetable on Monday at 10 am. Around each stop is a circle with a radius of roughly 450 meters indicating a comfortable 5-6 minute walk. In other words within a set time of 30 minutes, your trip is confined to around the stops of the one bus route. This is in a world where the buses aren’t as frequent and the coverage as a result suffers.

Coverage of all busses and trams in 30 minutes starting at Prins Bernhardlaan.

Coverage of all busses and trams in 30 minutes starting at Prins Bernhardlaan.

Now, take all the other buses and trams operated by U-OV and their timetables on a Monday morning at 10 am. Highlighted in orange are all the possible places you could travel to within 30 minutes from Prins Bernhardlaan. It’s because of the frequent service the opportunities to transfer are increased at stops used by other routes. A much larger portion of the region is now accessible in just 30 minutes and a 450-meter radius around each stop.1 The maps especially highlights the center of Utrecht, and Kanaleneiland to the south-west of the center and Overvecht to the north-east. This map does not include any of transfer opportunities presented by the multiple train stations in the area. For that we have the next map.

Coverage of all busses, trams, and trains in 30 minutes starting at Prins Bernhardlaan.

Coverage of all busses, trams, and trains in 30 minutes starting at Prins Bernhardlaan.

A similar isochrone map generated by CommuteTimeMap verifies the map I stitched together myself. However, theirs includes the trains. We can see places like Bunnik, Driebergen-Zeist, Houten, Houten Castellum and others all light up, which are accessible by train. This comes to show just how much more of the region is accessible with well-planned routes and a frequent schedule.2

This map could be filled out even further, if you taken into account Utrecht reputation for being a bicycle friendly city like most in the Netherlands. Starting your journey on a bicycle to a nearby bus hub or train station could potentially allow you to travel even further within the same 30 minute time span.


  1. I’m actually pleased to see the map from CommuteTimeMap somewhat aligns with the one I scraped together with half broken Python scripts and QGIS. Once I get a better grip of parsing GTFS data, I plan to revisit this and create a more accurate map exhausting each route. The map I created is still accurate and I made sure to verify the extremities manually through 9292.nl ↩︎

  2. CommuteTimeMap also may be using a wider radius at each stio than the 450 meters I used to symbolize a 5-6 minute walk. ↩︎


Related ramblings