Urban Studies Thesis: Bike Sharing Systems Pt. 3 – The Pricing Structure as Evidence for Separated Bike Lanes

This is the 3rd installment of a series of posts describing my thesis for the SFU Master of Urban Studies program.  My first two thesis articles here and here examined some of the underlying reasons for my research.  Basically our cities have been built almost to solely accommodate the automobile over the last century.  This has created a deficit in equitable high quality options for getting around the city.  One way to introduce equity into our cities, is to level the playing field by making it safe and easy for everyday people aged 8 to 80 years old to walk, cycle and use transit.  A good litmus test is to ask yourself if you would send your grandma to walk or cycle to her doctor appointment or your child to walk or cycle to school alone.  If you answer no, then it is most likely that it is not safe to walk or cycle.

Since it is well known that separated cycling lanes help support cycling growth among those 60% of the population that are “interested but concerned” in cycling more.  It is also well known that transit agency suffer from the weak link known as the “first and last mile.”  Essentially the requirement to walk far distances to get to transit from home is a significant impediment for increasing transit ridership.

This study takes actual Washington D.C. bikeshare trip data (1.5 Million of them to be exact, that were made between April 1, 2013 and September 31, 2013) and attempts to determine if there is a relationship between where people were making trips and the built environment, nearby separated cycling lanes and nearby high frequency transit services.  It is the hope that this research will provide empirically based evidence showing how people are using bikeshare systems and provide strong support for the investment of separated cycling infrastructure lanes and the resulting safe cycling environments.  Developing the understanding of how people are cycling within cities and the relationship with other modes, will also help guide policy, implementation and design decisions for these bikeshare systems. There has been a recent explosion of bikeshare systems with over 600 systems worldwide, so this research would be useful.

world bike locations

My thesis uses a combination of visualization and statistical analysis to determine if there is a relationship between trips made using Washington D.C.’s Capital Bikeshare system.  There could be hundreds of factors that influence the rates of cycling, therefore this thesis only focuses on the strongest known factors influencing where people cycle.  This can be broken into the fourteen variables listed together which can be broken into two distinct groups describing the socio-economic demographics and the built environment surrounding each bikeshare station.  These fourteen variables were accounted for in the statistical analysis to determine if separated cycling infrastructure and high frequency transit services play a significant role influencing where people are making trips.

thesis variables

So what did my research find? One of the main findings was that the pricing structure of these systems heavily influence the trip patterns.  The Capital Bikeshare system is designed so that there are two main ways to access the system:

  1. Long term annual or monthly “subscriber” membership
  2. Short term daily or three day “casual” membership

A key feature about the Capital Bikeshare is that access to the system provides you unlimited trips for the duration of membership as long as trips are 30 minutes or less.

Capital Bikeshare pricing

When you look at the statistics of the 1.5 million trips analyzed in this study you find a strong tendency for most trips (88%) to be less than 30 minutes. When you look at the subscriber members you find this pattern strengthens with 97% of trips less than 30 minutes.  Casual users are a different story, 37% of casual user trips are over 30 minutes.  This indicates that based on the current operating model and pricing structure, the strongest revenue potential exists with casual members.

trip duration stats

It was found from a 2012 Buehler study that casual users were most likely to be using the system for tourism or personal reasons mainly in the National Mall area.

For those not familiar with Washington D.C. this map will give you some basic context.  The National Mall is a large area in the center of the city with museums, monuments and parks.

Washington D.C. Context map

My visualization research also found that the highest number of trips made by the casual users were also located within the National Mall area.

Casual Users

Which also corresponded to the longest average trip length per station.

trip durations

This area also corresponded with some the highest use stations within the system.

finalaveragestart

And the highest volume of trip connections between stations.

spiderdiagramnew

Clearly there are high number of trips being made within the National Mall area by casual users, but the question is why? Why are casual users, many who are domestic tourists from the United States making a high number of trips within the National Mall, often for the first time (according to Buehler)?

national mall bike lane example

It could be that the National Mall has a high total length of off-road separated bike lanes.  These bike lanes provide a journey almost completely separated from traffic to view many of the United States greatest attractions including the Washington Monument, the Lincoln Memorial and the Capitol.  These wide and generous bike lanes may provide a comfortable cycling environments such that a large number consider it safe enough to cycle.

My statistical analysis certainly supports this.  The following tables shows the multivariable linear regression findings taking into account the fourteen variables described above.  Any rows that have been highlighted in green indicate a positive statistically significant relationship, while red shows a negative relationship.  No colour indicates no statistically significant relationship was found.  This table shows a clear positive relationship with separated bike lanes.

casual linear regression

From this research one can clearly anticipate that the pricing structure will affect how long people will use the bikeshare system, more specifically with subscriber members.  Casual members present the highest revenue generation, and are making most of their trips within the National Mall’s network of separated cycling lanes.  This provides strong support for expanding the network of separated cycling lanes in order to extend the areas that casual users can reach safely using the capital bike share system.  This in turn may see a growth in revenues.  You may also have noticed there was a negative relationship with bus based frequent transit services.  An upcoming installment will explain this important result.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s