Assessment of Bicycle Service Areas around Transit Stations

Susan Perry at MinnPost cites my collaborator Hartwig Henry Hochmair’s work in: Want to get more cars off the road? Improve bicycling infrastructure around transit hubs.

So I was intrigued to come across a study this week that examined how far cyclists in three large U.S. metropolitan areas are willing to ride to catch a bus or train that will take them the rest of the way to work.

One of those metro areas was Minneapolis-St. Paul. The other two were Los Angeles and Atlanta.

The study, which used data collected through mass-transit ridership surveys, found that while only a small percentage of people in the three metro areas ride their bike to a bus or train to commute to work, those who do tend to cycle an average of three miles or less — one to two miles in Minneapolis-St. Paul and Atlanta and a slightly longer three miles in Los Angeles (probably because that city’s weather is more conducive to biking).

The full title of the paper is: Assessment of Bicycle Service Areas around Transit Stations. [The paper itself is behind a paywall, though I am sure the author would share.] The paper’s abstract says:

Mobility hubs are major transit access points and an integrated part of multi-modal transportation planning efforts. For the implementation of bicycle infrastructure improvements around mobility hubs a better understanding of bicycle access distances is needed. Using responses from on-board travel surveys in three U.S. metropolitan areas, this study found that median bicycle access distances to transit stations are within the buffer radii suggested for community hubs (1 mile) and gateway hubs (2 miles) in long-range transportation plans. Multiple regression analysis identified several street and transit network characteristics affecting bicycle access distance, which should be considered when planning infrastructure improvements.

Fresh Data Hints at How to Close Biking’s Gender Gap | Wired

Jessica Schoner’s dissertation work on Bicycling’s Gender Gap got written up in Wired

The researchers turned up three especially interesting findings. The first is that in single-bicyclist homes, men are roughly twice as likely as women to ride. But when you’ve got two or more cyclists living together, that gap disappears. That could be because living with a cyclist encourages people of any gender to starting biking, or because people who enjoy cycling end up in the same home through marriage or friendship. “I don’t know what direction causality goes,” Schoner says.

The second finding is that among people who rode at least once on the day they kept their travel diary, there is no gender gap when it comes to the number of trips taken that day. In other words, women who ride do so just as frequently as men. “This suggests,” Schoner and Lindsey write, “that much of the remaining gender gap can be attributed to a participation gap, not an intensity gap.”

Finally, the 2010 data shows that having kids doesn’t lead to people biking less. That’s a change: In 2000, a parent was only half as likely to be a cyclist as a non-parent. There’s no gender difference here, but because women bear the greater burden when it comes to childcare, it’s encouraging news for those working to shrink the gender gap. “The relationship between having children and bicycling is complex and unclear,” Schoner says, but “having children may be becoming less of a barrier to bicycling over time.”

Factors Associated with the Gender Gap in Bicycling Over Time

Recent working paper:

Bicycling has grown in popularity over the past decade, but the gap in rates of bicycling between men and women in the United States (US) persists. This paper uses regional travel behavior study data from the Minneapolis-St. Paul Metropolitan Region in 2000 and 2010 to measure and model the gender gap in bicycling over time.


High Bike
High Bike

Findings from a series of statistical tests show that in aggregate, women bike less than men, and that growth in bicycling has been slower for women than for men over the past decade. However, stratifying the sample shows that women who live with at least one other adult bicyclist participate in bicycling at an equal rate as men. Similarly, frequency of bicycle trips among people who participate in bicycling differed by gender only slightly in 2000, and not at all in 2010. Binary logistic modeling results show that several factors, such as age and trip purpose, are associated with different bicycling outcomes for men and women, but some commonly hypothesized explanations, such as having children, were declining in effect or altogether insignificant.


These findings and conclusions are important for practice and research because understanding the nuances of the gender gap, such as the apparent gap in participation but not in frequency or the contagion effect of living with a cyclist, is essential for targeting programs effectively. This paper also identifies several travel behavior data collection limitations that complicate studying the gender gap, and offers recommendations for further study.

Differences Between Walking and Bicycling Over Time: Implications for Performance Measurement

Recent working paper:

Schoner, J., Lindsey, G., and Levinson, D. (2014) Differences Between Walking and Bicycling Over Time:  Implications for Performance Measurement

Walking and Biking Mode Shares in summer 2001 vs. 2010
Walking and Biking Mode Shares in summer 2001 vs. 2010
  • Transportation policies and plans encourage non-motorized transportation and the establishment of performance measures to assess progress towards multi-modal system goals. Challenges in fostering walking and bicycling include the lack of data for measuring rates of walking and bicycling over time and differences in pedestrians and bicyclists and the trips they make. This paper analyzes travel behavior inventories conducted by the Metropolitan Council in the Minneapolis-St. Paul Metropolitan Area in 2001 and 2010 to illuminate differences walking and bicycling over time and illustrate the implications for performance measurement. We focus on the who, what, where, when, and why of non-motorized transportation: who pedestrians and bicyclists are, where they go and why, when they travel, and what factors are associated with the trips they make. Measured by summer mode share, walking and bicycling both increased during the decade, but the differences between the modes overshadow their similarities. Using descriptive statistics, hypothesis testing, and multinomial logistic models, we show that walkers are different than bicyclists, that walking trips are shorter and made for different purposes, that walking and bicycling trips differ seasonally, and that different factors are associated with the likelihoods of walking or bicycling. While the increase in mode share was greater for walking than bicycling, the percentage increase relative to 2001 share was greater for bicycling than walking. Both walking and bicycling remain mainly urban transportation options. Older age reduces the likelihood of biking trips more than walking trips, and biking remains gendered while walking is not. These differences call into question the common practice of treating nonmotorized transportation as a single mode. Managers can use these results to develop performance measures for tracking progress towards system goals in a way that addresses the unique and different needs of pedestrians and bicyclists.

The Missing Link: Bicycle Infrastructure Networks and Ridership in 74 US Cities.

Recently Published:

Seattle bike network
Seattle bike network

Abstract: Cities promote strong bicycle networks to support and encourage bicycle commuting. However, the application of network science to bicycle facilities is not very well studied. Previous work has found relationships between the amount of bicycle infrastructure in a city and aggregate bicycle ridership, and between microscopic network structure and individual tripmaking patterns. This study fills the missing link between these two bodies of literature by developing a standard methodology for measuring bicycle facility network quality at the macroscopic level and testing its association with bicycle commuting. Bicycle infrastructure maps were collected for 74 United States cities and systematically analyzed to evaluate their network structure. Linear regression models revealed that connectivity and directness are important factors in predicting bicycle commuting after controlling for demographic variables and the size of the city. These findings provide a framework for transportation planners and policymakers to evaluate their local bicycle facility networks and set regional priorities that support nonmotorized travel behavior, and for continued research on the structure and quality of bicycle infrastructure and behavior.
Keywords Bicycling · Travel Behavior · Networks

Houten – bike city

Houten plaats OpenTopo

Houten, building on an old settlement, was initially planned as a city of 30,000 people where local transportation by bicycle is prioritized. Constructed from the 1960s onward as a reliever for Utrecht, it is connected by a short rail line with two stops in the town. Cars cannot cross the town, but can circumnavigate on a ring road (see attached map). The industrial and commercial sector is in the southwest of town, with good highway access. Though there is a good balance of jobs and workers, most residents work outside the town and most workers commute in, which is not surprising given its good connection with the rest of the Randstad. The architecture and feel of the place is otherwise very familiar to anyone who has visited a planned US, French, or UK new town from the same era (without the single family homes, most of the buildings are townhouses or apartments). Some 89 photos, mostly of Houten, can be seen on Flickr.

We toured it on bike one afternoon during the WSTLUR conference. (You may spot famous transportation educators in the photos.) Thanks much to the local officials who gave us the tour. These are my observations:

1. The center of town is the main train station (which was recently rebuilt). The number of tracks were increased and the station was elevated so it was easier to cross east-west.

2014-06-25 at 15-16-47

2. Under the train station is an enormous bicycle parking facility: Fietstransferium.

2014-06-25 at 15-18-51

3. There are many bike paths through town. Small humps are used to discourage cars, which are prohibited, and motor scooters and mopeds, which are as well, but seem common.

2014-06-25 at 15-34-39

4. The best, most vibrant part of the town is the old town, indicating there is much planners need to learn about recreating places.

2014-06-25 at 16-04-41

5. There are some shared roads, though most prohibit motors officially.

2014-06-25 at 16-31-04

6. The newest part of town is centered on Castellium, inspired by a Roman town.

2014-06-25 at 16-35-09

2014-06-25 at 16-44-12

7. One development is inspired a Norwegian Fjord town

2014-06-25 at 16-49-48

Some of my colleagues felt the town too “sterile” which is the rap given to new towns, and especially suburbs, everywhere. I don’t know what people are looking for, hypodermic needles on the street? It is of course a suburb of Utrecht, so the core city functions – especially entertainment and culture, will agglomerate there, as cities are where the childless youth seek to find mates. To conduct pop psychology and apply two of the Big Five personality traits this is a classic case of a trading off Openness to new ideas, which involves exposure to risk, and cities, and Neuroticism, which is fear based, and wants to minimize risk, and seeks more controlled environments (loosely, planned communities or suburbs), which I suspect at some level is in part correlated with age and parenthood.


I recently spent a week in Delft for the WSTLUR conference. My visits to Rotterdam and Houten are detailed in other posts.

As a place to consider the relationship between transport and land use (the mission of both WSTLUR and, Delft provides an ideal place that should be used as a model for emulation by planners.

Like most travelers I arrived in Delft via train (from Schiphol Airport). Aside from payment issues that Americans face due to lack of PIN and Chip (which will be rectified in 2015), train is extremely convenient, running on a frequent intercity schedule, even at 6:30 am on a Sunday morning, even with works being undertaken. It is a little bit confusing for the non-Dutch speaker, especially when certain trains don’t follow the printed schedule, but the electronic message boards will state which trains go to which destinations. However while the platform is usually given, the track may be dynamic, so pay attention. In any case, you can just ask a native (they are taller than you), almost all of whom speak English.

2014-06-22 at 06-35-12

On the train it is even simpler.

2014-06-22 at 07-28-23

Delft has an old railroad station that has seen better days. It is near the Tram Line, but due to works, the exit path is a bit circuitous.

2014-06-22 at 07-58-47

The new train station will be open soon, I am not clear if they are adding tracks.

2014-06-22 at 08-00-51

Lots of bikes park at the station.

2014-06-22 at 07-59-34

Along just about any street of importances, where bikes are cars don’t share space, bikes have dedicated, separated cycle tracks, lanes parallel to, but separated from, the adjacent motor vehicle lanes. These cycletracks, and the bike trails in general, are also apparently open to mopeds and motor scooters. Sometimes it is separated by short concrete barriers that allow water to flow. These appear to be retrofits.

2014-06-22 at 08-02-14

The separation is often done by vertical rather than horizontal spacing, that is, the cycle track is immediately adjacent to the roadway, but elevated maybe 10 cm.

2014-06-24 at 08-05-43

Cars are allowed for residents, but retractable bollards keep non-residents and unauthorized vehicles from driving and parking on local streets. (Some tour buses seem permitted as well).

2014-06-22 at 14-54-48

The canals define the city.

2014-06-22 at 08-06-47

There are pedestrian/bike/motorbike tunnels that are widely used despite US fears of unsafety and graffiti.

2014-06-22 at 13-03-15

The tunnels themselves may be below sea level

2014-06-22 at 15-35-25

The town square (the Markt in Centrum) is programmed with lots of activities on the weekend. On Sunday it was innovative environmental technologies …

2014-06-22 at 14-31-55

and acrobats.

2014-06-22 at 14-32-17

Another square (De Beestenmarkt, where various beasts were once traded) is home to outdoor drinking and dining.

2014-06-25 at 19-16-18

Saturday is market day.

2014-06-28 at 10-57-26

The public transport buses in town are contracted out to a private company Veolia.

2014-06-22 at 14-45-18

2014-06-22 at 14-52-15

In many places, there are shared spaces, including both commercial districts,

2014-06-22 at 14-51-01

along the canals,

2014-06-22 at 14-50-58

and residential neighborhoods.

2014-06-22 at 08-12-34

2014-06-22 at 08-13-55

Delft is a tourist destination in its own right, and adjacent to a major university: TU Delft, thereby generating additional demand.
The campus of TU Delft is getting a tram, but in the meantime relies on bus.

2014-06-24 at 08-12-25

In short, it is beautiful. The buildings are (mostly) beautiful. The canals are beautiful. The public squares are beautiful. The cobblestone streets are beautiful. The massing and scale is beautiful. Also, it is entirely walkable or bikeable.

My Flickr sets are here 1, 2, 3, 4, and 5.

Access by Nice Ride |

Cross-posted at Access by Nice Ride

Access by Nice Ride

I took my first Nice Ride last week. I had ridden a bicycle before, and I had used bike rental before (though not one of the modern automated systems), and I had used transportation vehicle “sharing” before (see e.g. ZipCar and Car2Go with mixed results), but I had not put all those things together. I finally figured since I am supervising research on this, I should actually become a member and test the system first person. [Not that this is a requirement, a medical researcher need not infect herself with a disease to study it, fortunately bike sharing is likely to be less lethal].After signing up online. My subscription key came quickly in the mail. My first ride worked technically well. I inserted my key, and got a bike, and rode (helmet-less) from all of Coffman Union to McNamara on the University of Minnesota campus (1 km), riding on the mostly car-less Washington Avenue Bicycle Mall most of the distance, and then found the station to return the bike to, and pushed it in, and saw a green light, and left. Check out was simpler than Car2Go.

For me personally, the functional markets of NiceRide seem limited at the moment, (see map … sadly without actual bicycle trails marked) there isn’t a station too near my house, nor just outside my office. This particular ride did not save me any time, but hey, it got me a blog post. Perhaps NiceRide will slightly shorten a trip to Dinkytown or WestBank. City-wide there are lots of Origin-Destination pairs where NiceRide would be useful.

As part of a research project (Nice Stations: An Exploration of Nice Ride Bike Share Accessibility and Station Choice) published earlier this year, Jessica Schoner and I systematically analyzed the accessibility differential created by NiceRide vs. Walking. We write.

Bike-share stations provided an increase in accessibility to jobs relative to walking at medium and high time thresholds. For short thresholds (e.g., 5 to 10 minutes), the cost of walking to a station to retrieve a bike consumed too much of the travel time budget, resulting in fewer jobs being accessible by Nice Ride than by walking directly. At 15 minutes, using Nice Ride provides access to 1.7 times as many jobs as walking on average in blocks that are within a 15-minute walk to a station. The peak advantage occurs at 30 minutes, where bike-share provides access to 221% more jobs than walking.

Nice Ride - Walk Accessibility at 40 minutes

The Figure shows where bike-share has the strongest advantage at the 40-minute threshold. Yellow and brown areas indicate higher job accessibility by bike-share than walking, and pink areas indicate the reverse. In downtown Minneapolis and immediately surrounding neighborhoods, bike-share improves job accessibility, but the areas are dense enough that walking still provides access to a large number of jobs. The dark brown ring shows the boundary where the utility of walking declines and bike-share remains high. Much of this area is lower density with fewer jobs, and it is too far from downtown for pedestrians reach it within the threshold. Bike-share’s higher travel speeds continue to enable people with access to a station to reach major job centers in and near downtown. Although downtown St. Paul also has a high concentration of jobs, the distribution of stations at the end of the 2011 season did not extend far enough to provide a benefit over walking.

Exploring Nice Ride job accessibility and station choice

Exploring Nice Ride job accessibility and station choice

Although bike share systems are becoming more popular across the United States, little is known about how people make decisions when integrating these systems into their daily travel. For example, when more than one bike share station is located nearby, how do users choose where to begin their trip, and what factors affect their decision?

Nice Ride station

In a study funded by CTS, researchers from the U of M’s civil engineering department sought to answer this question by investigating how people use the Nice Ride bike share system in Minneapolis and St. Paul. Professor David Levinson and graduate student Jessica Schoner examined how Nice Ride affects accessibility to jobs and developed a model to predict station choice.

In the first part of the study, the researchers created maps showing accessibility to jobs by census block for both Nice Ride and walking—as well as the difference between the two—at time thresholds ranging from 5 to 55 minutes. At lower thresholds, fewer census blocks have job accessibility via Nice Ride because of the time it takes for a person to walk to the Nice Ride station. However, at higher time thresholds, Nice Ride provides an improvement over walking. Overall, in blocks with both Nice Ride and walking job accessibility, Nice Ride provides access to 0.5 to 3.21 times as many jobs as walking.

In 2013, Nice Ride operated 170 stations with about 1550 total bikes in Minneapolis and St. Paul.

By comparing Nice Ride to walking, the study demonstrated that walking can successfully be used as a baseline to show how a bike share system improves job accessibility. The results also pinpointed when and where Nice Ride had the strongest accessibility advantage over walking.

“This type of information can be used by bike share system planners to identify where new stations could be built to maximize their impact on job accessibility,” Schoner says. “They could also look at accessibility to other destinations, like parks, grocery stores, or tourist attractions, depending on the goals of their system.”

Levinson and Schoner also developed a theoretical model for bike share station choice. The model considers users’ choice of a station based on their preference for the amount of time spent walking, deviation from the shortest path (the closest station may not be in the direct path of the person’s destination), and station amenities and neighborhood characteristics.

Nice Ride station

Findings show that people generally prefer to use stations that don’t require long detours to reach, but a station’s surroundings also play an important role. For example, stations located near a park and in neighborhoods with lower crime rates were more likely to be chosen as the starting point of a bike share trip. Results also show that commuters value shorter trips and tend to choose stations that minimize overall travel time, while users making non-work-related trips choose stations that allow them to spend more of their time biking, even if the total travel time is longer.

Understanding people’s station preference can help provide guidance to planners for bike share system expansion, densification, and optimization, Schoner says.

“For instance, even though spacing stations along a route would allow people to walk in the direction of their destination to pick up a bike, people’s strong preference to spend more time biking indicates that clustering stations near where they are starting and ending their trips might make more sense,” Schoner says.

Related Links