Exploring Shared-Use Mobility through Hubway Bikeshare

 

Project Manager

Michelle Scott

 

Project Principal

Annette Demchur

 

Data Analysts

Steven Andrews

Linghong Zou

 

Graphics

Steven Andrews

Jane Gillis

Kate Parker O’Toole

Linghong Zou

 

Cover Design

Kate Parker O’Toole

 

 

The preparation of this document was supported
by the Federal Transit Administration through
MassDOT 5303 contracts #88429 and #94643.

 

Central Transportation Planning Staff

Directed by the Boston Region Metropolitan

Planning Organization. The MPO is composed of

state and regional agencies and authorities, and

local governments.

 

 

March 2017


Hubway Service Area Map: This map shows the four municipalities--Boston, Brookline, Cambridge, and Somerville—that have Hubway service, and where they are located in the MPO region.


ABSTRACT

 

Bikesharing is an example of a shared-use mobility service, which involves sharing vehicles, bicycles, or other modes of travel, and provides users with short-term access to transportation. This report on Hubway, the bikesharing system that serves Boston, Brookline, Cambridge, and Somerville, includes three components:

 

  1. An analysis of Hubway member characteristics and travel behaviors, using survey data collected in February 2016. This information sheds light on how members use the system, including in conjunction with other modes. It explores not only member responses to individual questions, but also the relationships between different member characteristics and behaviors using cross-tabulations and a regression model. Survey results of members’ most recent trips show that many members used Hubway to meet essential transportation needs, such as commuting, and that many members selected Hubway because it was the fastest way to their destination. Top alternative modes for these most-recent trips include public transit (36 percent) and walking (33 percent).

 

Staff also used a statistical model to explore factors that might encourage Hubway members to replace motor vehicle trips with Hubway trips going forward. The model showed that respondents who had been Hubway members for longer periods, continuously or intermittently, had a greater propensity to replace motor vehicle trips with Hubway trips. The results of this survey analysis could be used to inform outreach to current or prospective users or enrollment campaigns.

 

  1. A discussion of general Hubway usage trends during 2015, according to season, user group, day of the week, time of day, and geographic location. This information provides a recent and multidimensional view of system usage, including variations in how Hubway members use the system compared to short-term, or casual, users.  

 

  1. An analysis of Hubway’s potential to complement or compete with MBTA fixed-route transit service for particular origin-destination (O-D) pairs. Staff used an open-source trip-planning tool to generate alternative transit itineraries for Hubway trips, and calculated a ratio for each trip to compare Hubway and transit travel times. Staff used this ratio, along with information about transit modes and transfers, to compare transit and bikesharing options for trips and O-D pairs. The tool recommended walking as an alternative for 31 percent of Hubway member trips, and bus as an alternative for another 36 percent of trips. Only 11 percent of the Hubway member trips analyzed would have been faster, or comparable in travel time, via transit. Staff also examined O-D pairs to identify where travel by Hubway may complement or compete with the MBTA fixed-route transit system.

This information could be used to inform policies, information campaigns, or investments with potential to enhance Hubway’s ability to support first-and-last-mile connections to transit, or otherwise offer a non-single-occupant vehicle alternative. This information also could help identify locations where transit could be improved to make it a more competitive option. The analytic techniques used in this report could be applied to analyses of other shared-use mobility datasets.

 


 

 

table of CONTENTS 

ABSTRACT

1          Hubway Bikesharing Research Overview

2          Hubway Member Survey Research

2.1         Member Survey Data Overview

2.2         Member Characteristics

2.3         Members’ Hubway Use

General Trip Information

Most Recent Trip Information

2.4         General Travel Characteristics of Members

Commute and Non-Commute Trips by Mode

Potential Connections to Transit

Auto Trips Replaced by Hubway Trips

2.5         Summary of Findings from the 2015 Hubway Member Survey

Choosing Bikesharing as a Travel Option

Reducing Motor Vehicle Dependence

3          Hubway Trip Data Research

3.1         Trip Data Source

3.2         General Trip Trends

Trips by Year and Season

Trip Data by User Type

Trip Data by Day of Week and Time of Day

3.3         Station-level Data Trends

3.4         Origin-Destination (O-D) Pair Data Trends

General O-D Pair Trends

O-D Pairs by User Type, Day of the Week, and Time Period

4          Hubway’s Potential to Complement or Compete with Transit

4.1         Comparing Hubway and Transit Trip Characteristics

4.2         Tools for Comparing Travel Times

4.3         Results for Individual Trips

4.4         Results for Origin-Destination Pairs

4.5         Identifying O-D Pairs that May Complement or Compete with Transit

4.6         Results for O-D Pairs with Large Shares of Trips that are Faster
or Comparable in Travel Time by Transit

4.7         Results for O-D Pairs with Large Shares of Trips that are Much Slower by Transit

4.8         Results for O-D Pairs with Large Shares of Walk-Only Trips

4.9         Summary of Findings

 

Table of Tables

TABLE 2-1 Numeric Values Assigned to Motor Vehicle Trip-Reduction Categories

TABLE 2-2 OLS Regression Results Measuring Hubway Members’  Propensity
to Replace Auto Vehicle Trips with Hubway Trips

TABLE 2-3 Motor-Vehicle Trip Replacement OLS Regression Model Statistics

TABLE 3-1 Average Hubway Trips per Day by User Type and Day of Week
(High-Activity Season)

TABLE 3-2 Top-Ten Stations by 2015 Station-Visit Volume (High-Activity Season)

TABLE 3-3 Weekday Time Periods

TABLE 3-4 O-D Pairs Categorized by Trip Volume

TABLE 3-5 Stations with High Volumes of “One-Station” Trips

TABLE 4-1 Travel-Time-Ratio Categories

TABLE 4-2 O-D Pairs by Percentage of Trips that Would Have Been Faster
or Comparable by Transit

TABLE 4-3 O-D Pairs by Percentage of Trips that  Would Have Been Much Slower
by Transit

TABLE 4-4 O-D Pairs by Percentage of Trips that Would Have Had Walk-Only Itineraries

TABLE 4-5 High Volume O-D Pairs with Trips Classified
by Travel-Time-Ratio Category

TABLE 4-6 Day-and-Time Periods

 

Table of Figures  

FIGURE 2-1 2015 Survey Respondents by Recent Membership Type

FIGURE 2-2 2015 Survey Respondents by Initial Membership Year

FIGURE 2-3 2015 Survey Respondents by Gender

FIGURE 2-4 2015 Survey Respondents by Race/Ethnicity

FIGURE 2-5 2015 Survey Respondents and  Population of Hubway Municipalities
by Age

FIGURE 2-6 2015 Survey Respondents and Population  of Hubway Municipalities
by Education Level

FIGURE 2-7 2015 Survey Respondents and  Population of Hubway Municipalities
by Household Income

FIGURE 2-8 2015 Survey Respondents by Reported  Access to Select Transportation Resources

FIGURE 2-9 2015 Survey Respondents by Biking Frequency

FIGURE 2-10 2015 Survey Respondents by Distance They Are Willing to Walk to
a Hubway Station

FIGURE 2-11 2015 Survey Respondents by Purpose and Starting Location  of Their Most Recent Hubway Trip

FIGURE 2-12 2015 Survey Respondents by Purpose  of Most Recent Hubway Trip
and Household Income

FIGURE 2-13 2015 Survey Respondents by Most-Preferred  Alternative Travel Mode for Their Most Recent Hubway Trip.

FIGURE 2-14 2015 Survey Respondents by Preferred Alternate  Motor-Vehicle Mode for Their Most Recent Hubway Trip.

FIGURE 2-15 2015 Survey Respondents by Preferred Alternate Mode for Their Most Recent Hubway Trip, and Household Income

FIGURE 2-16 2015 Survey Respondents by Most Recent Trip by Purpose  and Preferred Alternate Mode

FIGURE 2-17 2015 Survey Respondents by Their Primary Reason  for Using Hubway for Their Most Recent Trip

FIGURE 2-18 2015 Survey Respondents by Primary Reason for Choosing Hubway
for Their Most Recent Trip and Preferred Alternate Mode

FIGURE 2-19 Average Commute and Non-Commute Trips per Week by Mode and Bicycling Frequency Group

FIGURE 2-20 2015 Survey Respondents by Number of Hubway  Trips per Week Starting or Ending at Various Transit Locations

FIGURE 2-21 2015 Survey Respondents by Number of Trips  for which They Used Hubway Instead of Their Motor Vehicle.

FIGURE 3-1 Hubway Trips per Year

FIGURE 3-2 Available Hubway Stations per Year

FIGURE 3-3 Hubway Daily Trip Volumes in 2015

FIGURE 3-4 Hubway Daily Trip Volumes by User Type in 2015

FIGURE 3-5 Hubway Trip Volumes by User Type,  Day of Week, and Time of Day (High-Activity Season)

FIGURE 3-6 Hubway Station Locations and Visit Volumes

FIGURE 3-7 Top-Ten Stations by User Type, Day, and Time Period

FIGURE 3-8 Trip Volumes by Hubway O-D Pair

FIGURE 3-9 High-Volume O-D Pairs by User Type, Day, and Time Period

FIGURE 4-1 Alternatives to Hubway Trips by Travel-Time-Ratio Category

FIGURE 4-2 Alternatives to Hubway Trips by Mode

FIGURE 4-3 Transit Alternatives to Hubway Trips by Mode (Walk Trips Excluded)

FIGURE 4-4 Hubway Member Trips by Weekday Hour

FIGURE 4-5 Transit Alternatives to Hubway Trips  by Travel-Time Ratio Category and Weekday Hour

FIGURE 4-6 Hubway Trips by Alternative Recommended Modes by Weekday Hour

FIGURE 4-7 Hubway Trips by Alternative Recommended  Modes and Travel-Time-Ratio Category

FIGURE 4-8 Hubway Trips by Number of Transfers and  Travel-Time-Ratio      Category

FIGURE 4-9 Trips with Transit Itineraries by Mode and Number of Transfers

FIGURE 4-10 Trips with Transit Itineraries by Mode,  Number of Transfers, and Travel-Time-Ratio Category

FIGURE 4-11 O-D Pairs with 1,000 or More Hubway Trips by Primary Travel-Time-Ratio Category

FIGURE 4-12 O-D Pairs That May Complement or Compete with Transit (AM and PM Peak Periods)

FIGURE 4-13 O-D Pairs that May Compete with Transit (Weekday Off-Peak and Weekend Periods)

FIGURE 4-14 O-D Pairs with 75 Percent or More Trips Much Slower by Transit (Weekday AM and PM Peak Periods)

FIGURE 4-15 O-D Pairs with 75 Percent or More Trips Much Slower by Transit (Weekday Off-Peak and Weekend Periods)

FIGURE 4-16 Hubway Stations by Transit Modes Available within 200 Meters

FIGURE 4-17 O-D Pairs with 95 Percent or More Walk-Only Trips (Weekday AM and PM Peak Periods)

FIGURE 4-18 O-D Pairs with 95 Percent or More Walk-Only Trips (Weekday Off-Peak and Weekend Periods)

FIGURE 4-19 O-D Pairs with 95 Percent or More Walk-Only Trips by Connections to Rapid Transit (Weekday AM and PM Peak Periods)

FIGURE 4-20 O-D Pairs with 95 Percent or More Walk-Only Trips by Connections to Rapid Transit (Weekday Off-Peak and Weekend Periods)

FIGURE 4-21 O-D Pairs with 95 Percent or More Walk-Only Trips  by Number of Ends within 200 Meters of Rapid Transit Stations

 

Appendixes                                                              

Appendix A: Data Used in Analysis Stages

Appendix B: Station Names, Aliases, and Visit Data

Appendix C: Adjusting Transit Travel Time Estimates to Mitigate Bias

Appendix D: Methods used to Estimate Uncomfortable Conditions

Appendix E: Details on O-D Pairs with Large Shares of Trips Faster or Comparable
by Transit

Appendix F: Details on O-D Pairs with Large Shares of Trips Much Slower
by Transit

Appendix G: Details for O-D Pairs with Large Shares of Walk-Only Trips

 

1          Hubway Bikesharing Research Overview

This analysis of Hubway bikesharing user and trip characteristics is a component of a broader research initiative concerning shared-use mobility services operating in Greater Boston. Shared-use mobility services involve sharing vehicles, bicycles, and other travel modes so that users have short-term access to transportation on an as-needed basis.1 In particular, bikesharing services enable users to access bicycles on demand from bike docking stations or free-floating fleets around the service area.2 Shared-use mobility services such as bikesharing provide flexibility on an individual-trip level and may affect longer-term mode-share and car-ownership decisions. It is important to understand the role of shared-use mobility options in the region’s transportation system for short- and long-term transportation planning. 

 

In 2016, the Central Transportation Planning Staff (CTPS) of the Metropolitan Planning Staff (MPO) began work on a Massachusetts Department of Transportation (MassDOT)-funded study of shared-use mobility services in Greater Boston. The objectives of this study were to obtain data and provide analytical insights into

 

In addition to the Hubway analysis described in this report, CTPS conducted a literature review of shared-use mobility services, characteristics of their users, and impacts of these services on the transportation system.3 Because of limited available data on shared-use mobility services, CTPS focused on bikesharing when researching the study objectives.   

 

Hubway began operating in Boston in 2011 and expanded to Cambridge, Brookline, and Somerville in 2012.4 Financial support for the system has included grants from the Federal Transit Administration’s Bus Livability grant program and the Federal Highway Administration’s Congestion Mitigation and Air Quality CMAQ program. Funding also came from the Centers for Disease Control and Prevention, the Massachusetts Executive Office of Housing and Economic Development, the Boston Public Health Commission, the Barr Foundation, and corporate sponsorships from New Balance and other local private institutions.

 

The system is owned by the municipal governments of Boston, Brookline, Cambridge, and Somerville. Municipal funds have provided support for the Hubway system. The Metropolitan Area Planning Council (MAPC) oversees interaction and use of the system between municipalities. Motivate, formerly Alta Bicycle Share, is the contractor that designs, deploys, operates, and manages the Hubway system. As of the end of 2015, Hubway’s system footprint covered approximately 25 square miles and included 155 stations and 1,500 bikes. 5 The Hubway system features dedicated docking stations for its bicycles and automated credit card payment. Users can sign up for either an annual or a monthly membership or purchase a 24- or 72-hour pass.

 

The Hubway system presents a valuable case study for exploring the characteristics of shared-use mobility. Because the service has existed for several years, people living in Greater Boston have had an opportunity to integrate Hubway into their overall travel patterns, and potentially change their behaviors in response to its availability. Motivate publishes comprehensive Hubway trip logs on its website, which grants planners and researchers easy access to detailed data on individual trips from one Hubway station to another. Using trip data, along with Hubway member survey data obtained from Motivate via MAPC, staff conducted a three-part analysis:

 

  1. Analysis of 2015 Hubway member survey data, to gain insight into Hubway user characteristics and how and why they use bikesharing. This section describes what motivates members to use bikesharing, what kind of trips they make using bikesharing, and how bikesharing may fit into their overall travel behavior, including use of other modes. The analyses go beyond summary statistics of individual variables, and explore relationships between variables using cross-tabulations and a regression analysis. These results could be used to inform outreach or enrollment campaigns to help users (or potential users) to obtain or maintain their memberships.
  1. General analysis of Hubway trips taken during 2015. This section examines recent use of the Hubway system and describes when and where the system is being used, and how different customer segments (members or casual users) use the system differently.
  1. Analysis that compares Hubway trips with transit alternatives, and identifies cases where Hubway may complement or compete with the fixed-route transit system. Staff tested a technique using Open Trip Planner (OTP) that generated alternative transit and walking itineraries for Hubway origin and destination (O-D) pairs, and examined how these transit options would compare to Hubway trips in terms of travel time. This method could potentially be replicated with data from other transportation modes, particularly other shared-use mobility services. These results build upon the general analysis of Hubway trips and cite O-D pairs where OTP generally recommended a transit alternative that was faster or comparable in travel time to Hubway or much slower than Hubway. They also identified O-D pairs where OTP generally recommended that people walk. This information could be used to inform policies, information campaigns, or investments—such as integrated fare payment—that enhance Hubway’s ability to support first-and-last mile connections to transit or offer a non-single-occupant vehicle (SOV) option where it might not be feasible or desirable to enhance transit service, or where service could be crowded. This information could also be used to identify locations where transit could be improved to make it a more competitive option.  

2          Hubway Member Survey Research

2.1      Member Survey Data Overview

As of the end of 2015, Hubway had 13,248 annual members.6 In February 2016, Hubway issued a 2015 year-end survey to members who had an active annual membership at the end of 2015, to which 1,460 members responded. The survey asked questions on topics that included

This analysis focuses on the subset of survey questions that pertain most directly to the research objectives discussed in Part 1. Because this survey is specific to Hubway members, it does not provide insight into the characteristics and preferences of short-term-pass-purchasers. As discussed in the sections to follow, casual users took approximately 22 percent of Hubway trips in 2015; and differ from Hubway members in terms of when and where they make trips.

 

The survey form was distributed by email and data was collected during the winter of 2016, when people likely would modify their bicycling usage in response to the weather. Not all respondents answered all questions. In addition, responses reflect only those members who elected to respond; hence, these results may not be representative of all Hubway members. In spite of these limitations, this information still provides valuable insights into the travel behaviors of bikeshare users in greater Boston.

 

2.2      Member Characteristics

Figures 2-1 through 2-8 summarize the membership, demographic, and socioeconomic characteristics of Hubway survey respondents.

 

FIGURE 2-1
2015 Survey Respondents by Recent Membership Type

FIGURE 2-1: 2015 Survey Respondents by Recent Membership Type: This chart shows the distribution of survey respondents by recent membership type (annual, corporate/university, monthly, subsidized/low-income).

Data source: 2015 Hubway member survey. Sample size = 1,315.

 

Annual memberships and corporate or university-supported memberships are most common among the sample group, together comprising 93 percent of all reported memberships. Corporate or university-supported memberships are those where an organization collaborates with Hubway and covers some or all of members’ costs. 7

 

FIGURE 2-2
2015 Survey Respondents by Initial Membership Year

FIGURE 2-2: 2015 Survey Respondents by Initial Membership Year: This chart shows the distribution of survey respondents by when they joined Hubway.

Data source: 2015 Hubway member survey (sample size = 1,315).

 

Many respondents became Hubway members relatively recently; approximately half report joining in 2014 or 2015.

 

As shown in Figures 2-3 and 2-4, approximately two-thirds of respondents identified themselves as male, and 85 percent of respondents identified as white.

 

FIGURE 2-3
2015 Survey Respondents by Gender

FIGURE 2-3: 2015 Survey Respondents by Gender: This chart shows the distribution of survey respondents by reported gender.

Data source: 2015 Hubway member survey (sample size = 1,252).

Note: Those who identified as Other comprised less than one percent of respondents and are not shown above.

 

FIGURE 2-4
2015 Survey Respondents by Race/Ethnicity

FIGURE 2-4: 2015 Survey Respondents by Race/Ethnicity: This chart shows the distribution of survey respondents by reported race and/or ethnicity.

Data source: 2015 Hubway member survey (sample size = 1,190).

Notes: 1) The Multi category reflects those who selected more than one response option. 2) Other responses (not shown) includes American Indian and Alaskan Native (one person), Native Hawaiian and Pacific Islander (one person), and Other (three people). 3) The Other category includes those identifying as being of Middle Eastern ethnicity. 

 

Figures 2-5, 2-6, and 2-7 describe respondents’ age, household income, and education level. Staff aggregated responses to these questions in ways that made it possible to compare this information to US Census data on age, income, and education level for the Hubway service area population. 

 

FIGURE 2-5
2015 Survey Respondents and
Population of Hubway Municipalities by Age

FIGURE 2-5: 2015 Survey Respondents and Population of Hubway Municipalities by Age: This chart shows the distribution of survey respondents by their reported age, as well as the distribution of the population aged 18 and over in Boston, Brookline, Cambridge, and Somerville by age.

Data sources: 2015 Hubway Member Survey (sample size = 1,234), and 2010-14 American Community Survey (ACS) data for Boston, Brookline, Cambridge, and Somerville.

Note: A person must be at least 16 years of age to become a Hubway member. 

 

Of the respondents who answered this question, most were between 30 and 49 years old. The shares of respondents in the 18-to-29-year-old and 50-to-64-year-old categories are similar. As shown above, the largest share of survey respondents falls into the 30-to-49-year-old category, while the largest share of the 18-year-or-older population of the municipalities that have Hubway stations—Boston, Brookline, Cambridge, and Somerville—falls into the 18-to-29-year-old category.


 

FIGURE 2-6
2015 Survey Respondents and Population
 of Hubway Municipalities by Education Level

FIGURE 2-6: 2015 Survey Respondents and Population of Hubway Municipalities by Education Level: This chart shows the distribution of survey respondents by their education level, as well as the distribution of the population aged 18 and over in Boston, Brookline, Cambridge, and Somerville by education level.

Data sources: 2015 Hubway Member Survey (sample size = 1,263), and 2010-14 ACS data.

GED = general education development (high school equivalency).

 

The survey respondent group also has higher educational levels compared to the 18-year-or-older population in the Hubway service area municipalities. As shown above, 94 percent of survey respondents earned a bachelor’s degree or higher, compared to 46 percent of the service area population.   

 


 

FIGURE 2-7
2015 Survey Respondents and
 Population of Hubway Municipalities by Household Income

FIGURE 2-7: 2015 Survey Respondents and Population of Hubway Municipalities by Household Income: This chart shows the distribution of survey respondents by their household income level, as well as the distribution of the population aged 18 and over in Boston, Brookline, Cambridge, and Somerville by household income education level.

Data sources: 2015 Hubway Member Survey (sample size = 1,086) and the 2010-14 ACS Survey Public Use Microsample (PUMS).8

 

The survey respondent group skews toward higher household incomes than the 18-year-or-older population in the Hubway service area municipalities. The two groups are similar in terms of their shares of people with incomes between $50,000 and $74,999 and between $75,000 and $99,999.

 

The characteristics of the 2015 Hubway member survey respondents generally correspond to findings about the characteristics of bikeshare users. Fishman’s 2016 review of bikesharing literature reports that bikeshare users are more likely to be white (in the US and London), and male, and to have higher-than-average incomes and levels of education.9

 

FIGURE 2-8
2015 Survey Respondents by Reported
Access to Select Transportation Resources

FIGURE 2-8: 2015 Survey Respondents by Reported Access to Select Transportation Resources: This chart shows the percentage of survey respondents who report access to various transportation resources, including motorcycles, working and non-working bikes, carsharing services, cars/trucks/SUVs, and monthly MBTA passes/CharlieCards.

Data source: 2015 Hubway member survey (sample size = 1,327).

Notes: 1) 133 respondents who did not identify any transportation resources, and who did not answer subsequent questions, were excluded. 2) One person did not select “None of the Above” but did not identify any transportation resources.

SUV =  sport utility vehicle.

 

When asked about the transportation resources available to them, approximately 81 percent of respondents reported that they had access to a MBTA “Monthly Pass/Charlie Card, as shown in Figure 2-8.”10 Sixty-eight percent said they had access to a personal vehicle, while 63 percent reported access to a working bike.11 This information, combined with information about Hubway member incomes and education level, suggests that Hubway members generally are aware of, and have access to, a variety of transportation options; which enhances their ability to select the optimal mode for a given trip.

2.3      Members’ Hubway Use 

General Trip Information

Figures 2-9 and 2-10 provide information about members’ biking frequency during periods of good weather and the distance they are willing to walk to access a Hubway station.

 

FIGURE 2-9
2015 Survey Respondents by Biking Frequency

FIGURE 2-9: 2015 Survey Respondents by Biking Frequency: This chart shows the distribution of survey respondents by how often they ride a bike.

Data source: 2015 Hubway member survey (sample size = 1,397).

Note: Values sum to greater than 100 percent because of rounding.

 

FIGURE 2-10
2015 Survey Respondents by Distance
They Are Willing to Walk to a Hubway Station

FIGURE 2-10: 2015 Survey Respondents by Distance They Are Willing to Walk to a Hubway Station: This chart shows the distribution of survey respondents by how far, or how much time they are willing to spend, walking to a Hubway station.

Data source: 2015 Hubway member survey (sample size = 1,309).

 

The majority of survey respondents (88 percent) reported using Hubway or their personal bike during periods of good weather at least once a week, as shown in Figure 2-9. Figure 2-10 shows that 53 percent of respondents reported that they would be willing to walk as much as one-quarter mile to reach a Hubway station, although 16 percent said that the station would need to be “right here,” or within a three-minute walk.

 

Most Recent Trip Information

The member survey includes questions about respondents’ most recent Hubway trip. Figures 2-11 through 2-17 display the results of a subset of these questions.

 

FIGURE 2-11
2015 Survey Respondents by Purpose and Starting Location
of Their Most Recent Hubway Trip

FIGURE 2-11: 2015 Survey Respondents by Purpose and Starting Location of Their Most Recent Hubway Trip: This chart categorizes survey respondents’ most recent Hubway trip by its purpose and whether or not it was based at their home. Trip purposes include travel for work or school, for work or school related meetings, to/from public transportation, to/from errands or shopping, errands or shopping, social events or restaurants, and recreation or fitness, or “other.”

Data source: 2015 Hubway member survey (sample size = 1,426).

Note: Percent values in data labels reflect each trip purposes's share of total responses.

 

The top Hubway trip purpose identified by survey respondents was commuting to or from work or school (49 percent of all responses), followed by traveling to or from work-related meetings or public transportation or for errands. In his review of bikesharing systems in the US and internationally, Fishman found that commuting was a top trip purpose among bikesharing members.12 Fifty-five percent of respondents’ most recent trips were home-based. Figure 2-11 shows that less than half of trips to or from work-related meetings (12 percent of all trips) or public transit (12 percent of all trips) began from respondents’ homes. 


 

Figure 2-12 examines Hubway trip purposes with respect to respondents’ household income.

 

FIGURE 2-12
2015 Survey Respondents by Purpose
 of Most Recent Hubway Trip and Household Income

FIGURE 2-12: 2015 Survey Respondents by Purpose of Most Recent Hubway Trip and Household Income: This chart shows a cross-tabulation of survey respondents by their most recent trip purpose and their income level.

Data source: 2015 Hubway member survey (sample size = 1,084 respondents).

Notes: 1) The sample size for each income category is included in its data label. 2) The percent values in each column may not sum to 100 percent because of rounding.

 

More than 50 percent of the people in each income group reported making their most recent trip for work or school purposes (including meetings). Those who earned less than $35,000 made a larger share of Hubway trips for errands or shopping than other groups, while those with incomes in the $75,000–$99,999 range made a larger share of trips to connect with public transportation.


 

Figure 2-13 shows survey respondents’ most-preferred alternative travel modes had Hubway not been available for their most recent trip.

 

FIGURE 2-13
2015 Survey Respondents by Most-Preferred
Alternative Travel Mode for Their Most Recent Hubway Trip

FIGURE 2-13: 2015 Survey Respondents by Most-Preferred Alternative Travel Mode for Their Most Recent Hubway Trip: This chart shows the distribution of survey respondents by their most preferred alternative travel mode for their most recent Hubway trip. Options include public transit, walking, a combination of other modes not including motor vehicles, a motor vehicle, a combination of modes including a motor vehicle, a personal bike, not taking the trip, or “other.”

Data source: 2015 Hubway member survey (sample size = 1,430).

 

Thirty-six percent of respondents identified public transit as their most-preferred alternative for their recent trip, while 33 percent selected walking. Again, these results might have been affected by the survey being distributed in winter. However, these results do reflect findings from other bikeshare surveys across the country.13 In particular, the Shared Use Mobility Center (SUMC) found that respondents to their 2015 multi-city survey tended to cluster into active-transportation or motor vehicle-oriented groups when selecting alternatives to their preferred shared-use mode.14 Those who are inclined to use bikesharing might be most inclined to use transit or walking as a next-best alternative.

 

Figure 2-14 provides additional detail on the subset of respondents who selected “motor vehicle,” or “combination of modes including motor vehicle,” as their preferred alternate mode to Hubway for their most recent trip.

 

FIGURE 2-14
2015 Survey Respondents by Preferred Alternate
Motor-Vehicle Mode for Their Most Recent Hubway Trip

FIGURE 2-14: 2015 Survey Respondents by Preferred Alternate Motor-Vehicle Mode for Their Most Recent Hubway Trip: This chart shows, for those survey respondents who selected a motor-vehicle option as their preferred alternative to their most recent Hubway trip, their desired motor vehicle type. Options include taxi/Uber/Lyft, a personal car, Zipcar or Other Carshare, carpooling, or other.

Data source: 2015 Hubway member survey (sample size = 214).

Note: The sample for this chart only includes respondents who selected “motor vehicle” or “combination of modes including motor vehicle” as their preferred alternative to Hubway.

 

Respondents in this subset more frequently selected taxis or ridesourcing (56 percent) compared to personal vehicles (36 percent). This seems reasonable given that many respondents’ (41 percent) most recent trips started somewhere other than home.


Figure 2-15 examines respondents’ preferred alternate mode for their most recent Hubway trip with respect to their household income.

 

FIGURE 2-15
2015 Survey Respondents by Preferred
Alternate Mode for Their Most Recent Hubway Trip, and Household Income

FIGURE 2-15: 2015 Survey Respondents by Preferred Alternate Mode for Their Most Recent Hubway Trip, and Household Income: This chart shows a cross-tabulation of survey respondents by their preferred alternative mode for their most recent trip and their income level.

Data source: 2015 Hubway member survey (sample size = 1).

Notes: 1) The sample size for each income category is included in its data label. 2) The percent values in each column may not sum to 100 percent because of rounding.  

 

Across all income groups, walk and public transit make up the largest shares in terms of the modes respondents would have used if Hubway were not available. Those with household incomes greater than $150,000 had the highest share of those who selected motor vehicle as their alternate travel mode (13 percent). 


 

Figure 2-16 shows the percent of respondents who selected each alternate mode for a given trip purpose.

 

FIGURE 2-16
2015 Survey Respondents by Most Recent Trip by Purpose
 and Preferred Alternate Mode

FIGURE 2-16: 2015 Survey Respondents by Most Recent Trip by Purpose and Preferred Alternate Mode: This chart shows a cross-tabulation of survey respondents by the purpose of their most recent trip and their preferred alternative mode for that trip.

Data source: 2015 Hubway member survey (sample size = 1,424).

Notes: 1) The sample size for each trip purpose category is included in its data label. 2) The percent values in each column may not sum to 100 percent because of rounding.  

 

Public transit and walking are generally the travel alternatives respondents chose, although the rank of these alternatives (in terms of preference) varies depending on the trip purpose. Motor vehicles are a more popular alternative for social trips. Recreation or fitness trips had the highest share of respondents who said they would not have taken the trip if Hubway had not been available.


 

Respondents were asked about their primary reason for using Hubway for their most recent trip; their responses are shown in in Figure 2-17.

 

FIGURE 2-17
2015 Survey Respondents by Their Primary Reason
for Using Hubway for Their Most Recent Trip

FIGURE 2-17: 2015 Survey Respondents by Their Primary Reason for Using Hubway for Their Most Recent Trip: This chart shows the distribution of survey respondents by the reason they used Hubway for their most recent Hubway trip.

Data source: 2015 Hubway member survey (sample size = 1,412).

Note: The “One-Way Trip” category includes “One-Way T” responses.

 

Approximately two-thirds of respondents selected “fastest way to [their] destination” as their primary reason for using Hubway for their most recent trip, which is consistent with findings in bikesharing literature.15 However, other choices may have been more prominent had the survey been issued during a warmer season. 


 

Figure 2-18 compares the primary reason that members used Hubway for their most recent trip with the alternate mode they would have chosen for that trip.

 

FIGURE 2-18
2015 Survey Respondents by Primary Reason for Choosing Hubway
 for Their Most Recent Trip and Preferred Alternate Mode

FIGURE 2-18: 2015 Survey Respondents by Primary Reason for Choosing Hubway for Their Most Recent Trip and Preferred Alternate Mode: This chart shows a cross-tabulation of survey respondents by their preferred alternative mode for their most recent trip, and the reason they chose to use Hubway for that trip.

Data source: 2015 Hubway member survey (sample size = 1,407).

Notes: 1) The sample size for each alternate mode category is included in its data label. 2)  "Other" reasons for using Hubway include "reduce pollution/help the environment;" "Other;" "All;" "One-way Trip.” 3) The percent values in each column may not sum to 100 percent because of rounding.

 

The majority of respondents who otherwise would have taken public transit, taken a combination of modes, or walked said that their primary reason for using Hubway was that it was the “fastest way to [their] destination.” Approximately 18 percent of members who otherwise would have taken a motor vehicle said that they chose Hubway because it was the “least expensive option.”

 

2.4      General Travel Characteristics of Members

Commute and Non-Commute Trips by Mode

The Hubway member survey also offered insights into how Hubway fits into respondents’ overall travel habits. The survey asked respondents to report the number of one-way commuting trips and non-commuting trips they make during a typical week using various forms of transportation.16 CTPS used responses to these questions to calculate a weighted average number of commuting and non-commuting trips per week for each mode.17 Figure 2-19 shows these average numbers of trips for three groups of respondents: those who made four or more bicycle (Hubway or personal bike) trips per week, those who made between one-and-three bicycle trips per week, and those who made between one-and-three bicycle trips per month.

 

Figure 2-19 shows that those who reported biking four or more times a week made a much larger average number of Hubway commuting trips per week, compared to those who reported biking less frequently. Respondents in the “one-to-three trips per week” group made about one additional walking trip compared to the other two groups. Meanwhile, the “one-to-three trips per week” and “one-to-three trips per month” groups made at least one additional commute trip per week by auto compared to those who used Hubway or their personal bicycle more frequently. Those in the “one-to-three bicycle trips per month” group made slightly more transit commute trips per week, on average, than the other two groups, though there is not a large difference in the average number of transit-based commute trips across the three groups. The evidence that transit commute trips are relatively consistent across respondent groups while Hubway trips vary may suggest that some members in the “four or more bicycle trips per week” group use Hubway to connect to transit.   

 

For non-commute trips. walking and transit trip making across groups was fairly similar, although the “one-to-three bicycle trips per week” group made slightly more walking and transit trips, on average, than the “four or more bicycle trips per week” or the “one-to-three bicycle trips per month” groups. Overall, Hubway and transit appear to be prominent options for Hubway member commute trips, while auto is a more prominent option for non-commuting purposes. Walking is a prominent option for both types of trips.



FIGURE 2-19
Average Commute and Non-Commute Trips
 per Week by Mode and Bicycling Frequency Group

FIGURE 2-19: Average Commute and Non-Commute Trips by Mode and Bicycling Frequency Group: This chart includes two parts shown side-by side for comparison. The chart on the left shows weighted averages of the number of commuting trips taken by various modes per week for three groups of survey respondents: those who bicycle four or more times per week, those who bicycle one to three times per week, and those who bicycle one to three times per month. The chart on the right shows weighted averages of the number of non-commute trips taken by various modes per week for the same three groups of survey respondents.


Potential Connections to Transit

As mentioned in section 2.3, approximately 12 percent of Hubway respondents reported that they made their most recent trip to connect to transit. Later in the survey, respondents were asked to report how many of their Hubway trips started or ended near various public transit locations during a typical week. Figure 2-20 shows these results for subway/trolley, bus, and commuter rail.

 

FIGURE 2-20
2015 Survey Respondents by Number of Hubway
Trips per Week Starting or Ending at Various Transit Locations

FIGURE 2-20: 2015 Survey Respondents by Number of Hubway Trips per Week Starting or Ending at Various Transit Locations: This chart categorizes the shares of survey respondents who made certain numbers of trips per week that ended near access points for subway/trolley, commuter rail, or bus.

Data source: 2015 Hubway member survey (sample size (subway / trolley) = 1,166).

Notes: 1) Sample size (bus) = 970. Sample size (commuter rail) = 1,035. 2) Respondents who did not provide an answer for a specific mode were left out of that mode's sample.3) Percent values in each bar may not sum to 100 percent because of rounding.

 

As Figure 2-20 shows, more respondents reported making at least one trip per week that started or ended near a subway/trolley connection (59 percent), compared to trips that started or ended near bus or commuter rail (24 percent each). However, the wording of this question imposes some limitations on the information CTPS could glean about how frequently Hubway members may connect to transit stations, and what modes they use. The Hubway system is located in a relatively transit-dense area, so the likelihood of a Hubway station being near at least one transit option is high. While this question asks about the number of trips that started or ended near transit, it does not ask whether respondents used transit prior to or after making their Hubway trips, which would provide more information on how Hubway is used in multimodal trip making.

 

Auto Trips Replaced by Hubway Trips

The survey also asked respondents “If you have access to a motor vehicle, how many trips in a typical week in 2015 did you use Hubway instead of your motor vehicle?” This question could shed light on a great potential benefit of Hubway: encouraging users to shift away from single-occupant motor vehicle travel. Responses to this question are shown in Figure 2-21.

 

FIGURE 2-21
2015 Survey Respondents by Number of Trips
for which They Used Hubway Instead of Their Motor Vehicle

FIGURE 2-21: 2015 Survey Respondents by Number of Trips for which They Used Hubway Instead of Their Motor Vehicle: This chart shows the distribution of survey respondents by the number of motor vehicle trips they replaced with Hubway trips per week.

Data source: 2015 Hubway member survey (sample size = 1,152).

 

Approximately 55 percent of respondents reduced their motor vehicle travel by at least one trip per week. Research using bikesharing data from other cities, including Minneapolis/St. Paul (the Twin Cities), Montreal, and Washington DC similarly found reductions in driving among bikeshare members. 18

 

To explore this topic in more detail CTPS created an ordinary-least-squares (OLS) linear regression model to explore the factors that may influence respondents to replace motor vehicle trips with Hubway trips.19 This model estimated an individual’s tendency to replace vehicle trips with Hubway trips using this replacement ratio:

 

Replacement Ratio =

Number of Vehicle Trips Replaced by Hubway in a Typical Week/Number of Hubway Trips in a Typical Week

 

For this model, CTPS selected 600 respondents from the overall sample who reported

 

CTPS calculated the total number of Hubway trips that a respondent made each week by summing the Hubway commute and non-commute trips that the respondent reported, as discussed earlier in this section (see Figure 2-19). CTPS then examined the category the respondent selected for the number of motor vehicle trips she typically replaced with Hubway each week (see Figure 2-21). Because respondents did not provide raw values for the motor vehicle trips that they replaced each week, CTPS used the midpoint value as the representative value for the “1-3 trips”, “4-6 trips”, and “7-10 trips” categories, as shown in Table 2-1. For the “More than 10 trips” category, CTPS assumed a left-skewed distribution and therefore selected 11.5 trips as the representative value.

 

TABLE 2-1
Numeric Values Assigned to Motor Vehicle Trip-Reduction Categories

Motor Vehicle Trip-Reduction Category

Assigned Numerical Value

0 trips

0 trips

1-3 trips

2 trips

4-6 trips

5 trips

7-10 trips

8.5 trips

More than 10 trips

11.5 trips

 

CTPS examined different groups of variables from the survey that might influence an individual’s inclination to replace Hubway trips with motor vehicle trips, including but not limited to

 

CTPS removed survey responses that contained at least one missing value for these explanatory variables and ultimately used 462 Hubway member survey responses in the final model estimation. Table 2-2 describes the seven explanatory variables (including the model intercept20 ) that were ultimately included in the final model. All of these variables are statistically significant, meaning that it is highly probable that there is a relationship between each variable and the replacement ratio is not random.21

 

The model includes two explanatory variables that are difficult to explain directly: the home-based trip indicator and the Asian-respondent indicator. CTPS kept these statistically significant variables in the model because they might serve as proxies for a “true” factor that contributes to variations in the vehicle-trip replacement ratio. Meanwhile, the “years since first becoming a Hubway member” variable provides valuable information from a policy standpoint. If people can maintain their Hubway memberships over time—perhaps through financial support—they may be able to maintain the behaviors that would help them transition away from single-occupant motor vehicle use for some of their trips.


TABLE 2-2
OLS Regression Results Measuring Hubway Members’
Propensity to Replace Auto Vehicle Trips with Hubway Trips

Variable Description

Estimated Parameter

t-Statistic

Coefficient Interpretation

Description and Possible Explanation

Intercept

0.31

4.5

N/A

N/A

The share of automobile and taxi trips of all (commute and non-commute) trips in a typical week

0.37

3.4

A 10 percent increase in motor vehicle trip share results in a 3.7 percent average increase in a respondent’s motor vehicle trip replacement ratio.

Hubway members that rely primarily on motor vehicles are more likely to replace motor vehicle trips with Hubway trips (as opposed to those who may rely more on transit).

Home-based trip indicator (1 if the respondent's most recent Hubway trip was home-based, 0 if otherwise)

0.13

3.4

If a respondent's most recent Hubway trip was home-based, her replacement ratio increases by 13 percent, on average.

If a respondent’s most recent Hubway trip was home-based, we assume that she may be more likely to start her Hubway trips from home generally than would a member who reported that his most recent trip was not based at home. Using this assumption, members who typically make home-based Hubway trips may be more likely to replace motor-vehicle trips than if they were not typically making home-based trips. This makes sense if a member’s motor vehicle is parked near her home.

Commute trips by transit in a typical week

-0.01

-3.0

Each additional transit commute trip a respondent makes per week decreases his replacement ratio by one percent on average.

As members increase the number of commute trips they make by transit each week, they become less likely to replace motor vehicle trips with Hubway trips. This is probably because their Hubway trips would replace transit trips more often (as opposed to motor vehicle trips).

Asian respondent indicator (1 if the respondent is Asian, 0 if otherwise)

0.24

2.7

If a respondent is Asian, his replacement ratio increases by 24 percent on average.

This variable might be a proxy for where Hubway trips are being made. For example, Asian members may take Hubway trips in neighborhoods where taking trips by Hubway might be easier than taking trips by motor vehicle.

Years of Hubway membership

0.03

2.3

For each additional year that has passed since a respondent first became a Hubway member, her replacement ratio increases by three percent on average.

The more years that have passed since a person first became a Hubway member, the more likely she is to replace motor vehicle trips with Hubway trips.

Willingness to walk more than five minutes to a Hubway station (1 if yes, 0 if otherwise)

0.07

1.8

If a respondent is willing to walk longer than five minutes to access a Hubway station, his replacement ratio increases by seven percent on average.

Members that are willing to walk longer to reach a Hubway station might be more likely to make behavior changes that support a mode shift from motor vehicle to Hubway use.

Note: The “Descriptions and Possible Explanations” column suggests possible reasons behind the relationships between each explanatory variable and the response variable. However, this model can only highlight correlations between variables and not definitively explain why the relationship exists.  

N/A = Not available or applicable. OLS = ordinary-least-squares.


Table 2-3 shows the number of observations included in the final model, along with statistics that describe how well the model fits the data.

 

TABLE 2-3
Motor-Vehicle Trip Replacement OLS Regression Model Statistics

Model Parameter

Value

Number of observations

462

F-statistic

10.75

R-squared

0.124

Adjusted R-squared

0.113

Note: The p-value associated with the F-statistic is 0.000 or 3.396e-11

OLS = ordinary-least-squares.

 

The F-statistic for this model indicates that this model predicts changes in the response variable (motor-vehicle-trip replacement ratio) better than a model that included only the intercept, or the mean motor-vehicle replacement ratio value. In terms of the overall model fit, an r-square value of 0.124 means that the model explains 12 percent of the variability in the data. The model may have this relatively low R-squared value for the following reasons:

 

Though this model has a low R-squared value, which limits its ability to make precise predictions, it still provides valuable insights into the relationship between the variables, and the factors that may influence the extent to which a person might replace motor-vehicle trips with Hubway trips. The high t-statistics associated with the explanatory variables (see Table 2-2) indicate that these variables reveal significant trends in otherwise noisy and highly variable data.

 

2.5      Summary of Findings from the 2015 Hubway Member Survey

Of the research objectives mentioned in Section 1, the results of Hubway’s 2015 member survey provide insights into

Choosing Bikesharing as a Travel Option

In the 2015 Hubway member survey, the majority of respondents reported that their top reason for using Hubway for their most recent trip was that it was the fastest way to their destination. Figure 2-13 shows that this response far surpassed other reasons, including cost, health, or environmental benefits. Moreover, Figure 2-18 shows that this reason continued to be prominent regardless of the alternative mode the member might have taken for their trip. While not surprising, this finding speaks to the role that Hubway can play as part of an integrated transportation system. Figures 2-14 and 2-17 suggest that members may be making very small numbers of bikeshare trips that they would not otherwise have made, and that these trips were primarily for recreational, fitness, errands, or shopping purposes. It is important to note that because this research used member survey data, these results are biased towards individuals who have chosen to use Hubway. To conduct an unbiased study that examines people’s decisions to switch to bikesharing, one would need data from a broader population—including people who did not use Hubway—and comprised of individuals with the ability to choose from multiple modes for particular trips. This data then would describe how respondents would choose from various transportation modes to make particular trips.

 

The survey also provides information about who is taking advantage of the Hubway system to meet their travel needs. The results indicate that Hubway survey respondents tend to be better educated and have higher household incomes than the overall population of municipalities with Hubway service (Figures 2-3 and 2-7), suggesting that knowledge and resources may enable people to make the choice to use bikesharing. The City of Boston and the Boston Public Health Commission have collaborated to make subsidized memberships available to low-income individuals, and Hubway offers monthly memberships, which can increase access for those who might not be able to afford an annual membership.22 To the degree that these initiatives, along with public education campaigns, can be sustained and expanded, more people of various incomes and education levels may be able take advantage of the service. Hubway has also made new stations available in East Boston, Dorchester, Roxbury, and other areas during 2016. Therefore, the 2016 member survey possibly may reflect more diversity in respondent characteristics than it might have prior to this geographic expansion.

 

Reducing Motor Vehicle Dependence

Survey data on respondents’ most recent Hubway trips reveals mixed findings with respect to bike sharing’s ability to help individuals live a less car-dependent lifestyle. Many members made their most recent Hubway trip to commute, attend work or school meetings, access public transportation, or do errands, which suggests that people use Hubway to meet essential transportation needs (see Figure 2-11); thus, offering the potential for people to become less dependent on automobiles for regular or essential travel. Meanwhile, the majority of survey respondents reported that they otherwise would have walked or taken public transit for their most recent trip; and in cases where they would have used a vehicle, they more likely would have used a taxi or ridesourcing service (such as Uber or Lyft) than a personal vehicle.    

 

Bikesharing’s potential to reduce motor vehicle dependent lifestyles may be more apparent when one examines broader trip-making patterns. Figure 2-11 shows that more than half of respondents made their last Hubway trip for work or school, while Figure 2-19 shows that those who bicycle most frequently make a larger average number of commuting trips (an essential transportation purpose) using Hubway during a typical week. The survey results also show that 55 percent of respondents reported replacing at least one motor vehicle trip with a Hubway trip each week. The results of the regression model show that respondents who had been Hubway members for longer periods, continuously or intermittently, had higher propensities to replace motor vehicle trips with Hubway trips. This is a positive indicator that Hubway might help people to live a more car-free lifestyle. Future surveys could supplement these findings by adding questions about whether Hubway members have sold cars or postponed new car purchases because they had access to Hubway, or how they have changed their use of different modes since joining Hubway. Future research also could explore how members use Hubway differently based on their reported home zip codes.


3          Hubway Trip Data Research

While Hubway’s member survey data provide information about user characteristics, preferences, and self-reported behaviors, Hubway trip logs reveal when and where users actually make bikeshare trips. This makes it possible to look at trip-making behavior across all users and explore whether they are using the Hubway system in ways that potentially complement, or compete with, transit.

3.1      Trip Data Source

CTPS obtained trip logs for the years 2011 through 2015 from Motivate in February 2016, although Motivate also now provides trip logs on the Hubway website.23 These logs include trip duration (length of time), the start and stop dates and times of the trip, the start and end stations, and the bicycle identification number. The logs that CTPS obtained also list whether the person who made the trip was a Hubway member or casual user (that is, a short-term pass purchaser). If the person was a member, the data includes the member’s home zip code and gender. While this data is a rich resource for exploring shared-use mobility, it lacks several details that prevent it from telling the full story about a person’s trip:

In the absence of full details, CTPS has defined a Hubway trip as including only the time and activity between when a bicycle is checked out of a Hubway station dock and when it is returned to a station dock. In addition to Hubway trip logs, CTPS used data provided by Motivate about Hubway stations, including stations’ identification numbers, names, coordinates, and number of available bike docks.

 

3.2      General Trip Trends

Trips by Year and Season

Figures 3-1 and 3-2 below show the number of trips and stations that appeared in each year of Hubway data.24

 

FIGURE 3-1
Hubway Trips per Year

FIGURE 3-1: Hubway Trips per Year: This chart shows the number of Hubway trips taken per year from 2011 and 2015.

Data source: 2011-2015 Hubway trip logs.

 

 

FIGURE 3-2
Available Hubway Stations per Year

FIGURE 3-2: Available Hubway Stations per Year: This chart shows the number of stations available in the Hubway system each year from 2011 to 2015.

Data source: 2011-2015 Hubway trip logs.

 

 

Hubway launched service in Boston in 2011 and expanded into Cambridge, Somerville, and Brookline during summer 2012. During this time, total annual trips increased 276 percent, and total available stations increased by 75 percent.25 Total annual trips decreased by about five percent between 2014 and 2015, despite the increase in the number of stations. This likely can be explained in part by the heavy snowfall and extended periods of inclement weather during early 2015.

 

Subsequent components of this analysis focus on trips made during 2015. Figure 3-3 shows the distribution of Hubway trips over the course of 2015.

 

FIGURE 3-3
Hubway Daily Trip Volumes in 2015

FIGURE 3-3: Hubway Daily Trip Volumes in 2015: This chart shows the variation in the number of Hubway trips per month over the course of 2015.

Data source: 2015 Hubway trip logs.

 

Depending on the time of year, Hubway may have only a limited number of stations in operation, which contributes to the variations in Hubway usage over the course of the year. According to Motivate, 32 stations were open in Cambridge during winter and early spring 2015, during which the number of trips per day remained consistently below 1,000 until the full season opening on April 17, 2015.26 In the summer and fall months, the number of trips on the system generally remained between 4,000 and 6,000 per day and then began to decline in mid-November. Starting on November 25, system stations began to close; according to Motivate 111 stations were open through December 7, and 108 stations were open through the end of the year.27 After December 18, the number of trips per day trips generally remained below 1,000, similar to wintertime levels exhibited in the early parts of the year.

 

Trip Data by User Type

The Hubway 2015 trip data reflects trip made by both Hubway members and casual users of the Hubway system. According to Motivate, Hubway had 13,248 members and sold 102,445 24-or-72 hour casual passes during 2015; and system volumes reflect trips made by about 60,000 unique users.28 In 2015, casual users accounted for approximately 22 percent of the trips made on the system, while members accounted for the other 78 percent.

 

Figure 3-4 shows how the trips made by members and casual users are distributed throughout the year.

 

FIGURE 3-4
Hubway Daily Trip Volumes by User Type in 2015

FIGURE 3-4: Hubway Daily Trip Volumes by User Type in 2015: This chart shows the variation in the number of Hubway trips for Hubway members and casual users per month over the course of 2015.

Data source: 2015 Hubway trip logs.  

 

In general, the number of Hubway trips made by casual users per day is lower than the number of trips made by members, except during holidays and holiday weekends, when the number of casual user trips tends to exceed those made by members. This likely reflects that members are not making their typical commuting trips, while casual users are biking more for recreational purposes.

 

For the subsequent sections of this report, CTPS analyzed only Hubway trips within a “high-activity season,” the period between April 17, 2015 and December 18, 2015. CTPS analyzed this subset of trips to focus on the part of the year when the Hubway system was most widely available and most heavily used and to avoid data outliers that the wintertime trips may introduce, such as longer-than-typical durations because of inclement weather. To create this subset, CTPS removed approximately 45,000 of 1,122,000 trips or four percent of all 2015 trips, from the dataset, and retained approximately 1,077,000 trips for the next stage of analysis. CTPS excluded other subsets of trips to meet the needs of subsequent analyses, as we discuss later in this section and in Section 4. Appendix A includes a diagram that identifies each stage in the analysis when trips were excluded from the 2015 dataset. 

 

Trip Data by Day of Week and Time of Day

Table 3-1 shows the typical daily volume of trips made by members and casual users during the previously defined “high-activity” season (4/17/2015–12/18/2015).

 

TABLE 3-1
Average Hubway Trips per Day by User Type and Day of Week
 (High-Activity Season)

User Type

Number of
Weekday Trips

Number of
Weekend Trips

Casual User

844

1,296

Member

3,593

2,996

Overall Users

4,438

4,292

Data source: 2015 Hubway trip logs (4/17/2015–12/18/2015).

 

Members tend to make more trips per day than casual users on both weekdays and weekend days. Unlike members, casual users tend to make more trips on weekend days than on weekdays. 

 

Figure 3-5 shows the distribution of member- and casual-user trips made during the 2015 high-activity season according to the day of the week and time they were made.


 

FIGURE 3-5
Hubway Trip Volumes by User Type,
Day of Week, and Time of Day (High-Activity Season)

FIGURE 3-5: Hubway Trip Volumes by User Type, Day of Week, and Time of Day (High-Activity Season): This chart shows how the number of Hubway member and casual user trips varies over the course of the week. This chart reflects trips taken during the period from April 17, 2015 to December 18, 2015.

Data source: 2015 Hubway trip logs (4/17/2015–12/18/2015).

 

During weekdays, member trips exhibit clear peaks in usage: around 8:00 AM and 5:00 PM (commuting periods), while the number of casual-user trips gradually increases during the day to a peak at around 5:00 PM. On weekend days, members and casual users are similar in terms of the number and timing of trips they make, which are distributed gradually throughout the day and peak in the afternoon. These weekday and weekend trip distributions generally reflect usage patterns identified in previous analyses of Hubway trip making29 , and in similar research of bikeshare systems around the country.30  

 

3.3      Station-level Data Trends

Analyses at the station and origin-destination-pair (O-D pair) level place Hubway trip data in a geospatial context. The 2015 Hubway trip log referenced 156 stations. CTPS combined two stations near the Old State House into one record and two stations near TD Garden into one record because these stations were very close together, and established a set of 154 stations for the purposes of analysis.31 For this report, CTPS assigned aliases to replace the station names assigned by Hubway, which can be long. Hubway station names and their aliases are listed in Appendix B. 

 

Figure 3-6 displays these stations, which have been classified by their total station visits between April 17, 2015 and December 18, 2015. Station visits include trips originating or arriving at a given Hubway station, according to the trip logs. In general, stations with a large volume of visits are close to MBTA transit facilities—particularly along the Orange and Green Line subway lines in the northern part of Boston, and along the northern portion of the Red Line, extending from South Station north into Cambridge and Somerville. High-volume stations also are often near or within college and university campuses, such as Harvard, MIT, and Boston University.

 

Table 3-2 shows the top-10 Hubway stations by the number of station visits made during the high-activity season. These stations account for 20 percent of station visits. The top-four stations—TD Garden, South Station, MIT at Massachusetts Ave. and Amherst Street, and Harvard Square at Massachusetts Ave. and Dunster Street—account for 10 percent of all station visits, while 37 stations (24 percent of all stations) account for 50 percent of all station visits.

 

TABLE 3-2
Top-Ten Stations by 2015 Station-Visit Volume (High-Activity Season)

Station Name

Municipality

Station Visits in 2015a

TD Garden

Boston

65,070

South Station

Boston

61,372

MIT: Mass Ave

Cambridge

53,141

Harvard Square: Mass Ave @ Dunster St.

Cambridge

42,394

Boston Public Library

Boston

38,389

Central Square

Cambridge

37,614

Lafayette Square

Cambridge

35,395

MIT: Stata Center

Cambridge

34,545

Charles Circle

Boston

34,108

Back Bay / South End Station

Boston

32,110

Total Visits

null

434,287

Average of all 154 Stationsb

null

13,987

Data sources: 2015 Hubway trip logs (4/17/2015–12/18/2015).

a The number of bicycle docks varies by station. b This average excluded visits made to the Hubway Warehouse at 18 Dorrance Street, Boston (149 visits).

 

Figure 3-6 ranks stations by visit volumes and according to user type (member or casual user), day of the week (weekday or weekend day), and time of day. The top-10 stations in each category are shown on the chart, which also shows the stations that are prominent across multiple categories.

 

FIGURE 3-6
Hubway Station Locations and Visit Volumes

FIGURE 3-6: Hubway Station Location and Visit Volumes: This map shows the locations of Hubway stations and classifies them by visit volumes for the period from April 17, 2015 to December 18, 2015.

 

Table 3-3 describes the hours covered by each weekday time period; the weekend category includes all hours on Saturdays and Sundays.

 

TABLE 3-3
Weekday Time Periods

Time Period

Relevant Hours

Late Night / Early Morning

12:00 AM–6:59 AM

AM Peak

7:00 AM–10:00 AM

Midday

10:01 AM-3:59 PM

PM Peak

4:00 PM–7:00 PM

Evening / Night

7:01 PM–11:59 AM

Source: Central Transportation Planning Staff.

Note: These time periods are based on distribution of Hubway member trips over the course of a weekday.

 

Twenty-six of the 36 stations in Figure 3-7, below, fall within the top-30 stations, as ranked by overall station visit volumes between April 17 and December 18, 2015; these stations are shown in bold. However, these stations vary somewhat in terms of when they are used and by whom. Several stations rank high across all, or nearly all categories, such as TD Garden, Harvard Square, and MIT: Massachusetts Avenue at Amherst Street. Some stations—such as Boylston Street at A rlington Street, the Boston Public Library (Boylston Street), and the Esplanade—rank high in terms of casual user visits across all day and time categories, but do not rank as high in terms of member visits. Other stations, such as those at Back Bay and Central Square, experience more visits by members, particularly during commuting hours.

 

FIGURE 3-7
Top Ten Stations by User Type and Day-and-Time Period


FIGURE 3-7: Top-Ten Stations by User Type and Day-and-Type Period: This chart identifies the top ten Hubway stations by station visits in various categories, which are defined by user type (member or casual user), day type (weekday and weekend day), and weekday time period, if applicable. Station visits have been tabulated for the period from April 17, 2015 to December 18, 2015.


 

3.4      Origin-Destination (O-D) Pair Data Trends

General O-D Pair Trends

CTPS organized the trip data into origin-destination flows to explore where people have used Hubway to make connections. According to the 2015 Hubway trip logs, of the 1,077,000 trips during the high-activity period, 97 percent have different origin and destination stations. Seventy-nine percent of these trips were made by Hubway members, while the remaining 21 percent were made by casual users. These trips have been classified into 17,615 O-D pairs, which account for the direction of the trip.32 Table 3-4 classifies these O-D pairs by their trip volumes during the high-activity season.

 

TABLE 3-4
O-D Pairs Categorized by Trip Volume

Trip Volume per O-D Pair

Number of O-D  Pairs

Percent of All O-D Pairs

Cumulative Percent of All O-D Pairs

Number of Trips

Percent of All Trips

Cumulative Percent of All Trips

1,000 or more

36

0.2%

0.2%

55,537

5.3%

5.3%

500 to 999

201

1.1%

1.3%

134,708

12.9%

18.2%

200 to 499

1,033

5.9%

7.2%

306,067

29.3%

47.6%

100 to 199

1,684

9.6%

16.8%

232,809

22.3%

69.9%

Less than 100

14,661

83.2%

100.0%

314,166

30.1%

100.0%

Total

17,615

100.0%

100.0%

1,043,287

100.0%

100.0%

Data source: 2015 Hubway trip logs (4/17/2015–12/18/2015).

O-D = Origin-destination.

 

Table 3-5 shows that a relatively small share of O-D pairs covers a large share of Hubway trips. For example, 1.5 percent of O-D pairs had 500 or more trips, and these comprised 18.2 percent of trips on the system. Meanwhile, 83.2 percent of O-D pairs had less than 100 trips, but these pairs accounted for only slightly more than 30 percent of trips during the high-activity season. 

 

Figure 3-8 displays the O-D pairs with 500 or more trips or more during the high-activity season.

 

FIGURE 3-8
Trip Volumes by Hubway O-D Pairs

FIGURE 3-8: Trip Volumes by Hubway O-D Pair: This map shows Hubway trip origin-destination (O-D) pairs identified in Hubway trip data, categorized by trip volume. These origin-destination pairs are based on trips taken between April 17, 2015 and December 18, 2015.


 

High-volume flows exist between Hubway stations at Teele Square and the northern end of Linear Park and Davis Square, between and within Harvard and MIT, and on the link over the Charles River at the Harvard Bridge. Other high-volume flows exist between North Station (TD Garden) and South Station, between North Station and Congress/Sleeper Streets, between South Station and the Boston Waterfront, and between TD Garden and points in Charlestown. Woodruff identified similar high-volume flows in his analysis of Hubway trips between July 2011 and October 2012.33 These findings also correspond to a Hubway trip flow analysis conducted for the GoBoston 2030 Vision Framework, which highlights high-volume flows between North and South Stations and over the Mass Ave. (Harvard) bridge.34  

 

O-D Pairs by User Type, Day of the Week, and Time Period

Figure 3-9 shows the geographic distribution of O-D pairs by time period, and by the type of user. The “Weekend” maps show system use for the full day on weekend days.

 

The variations in trip activity shown in Figure 3-9 correspond to CTPS’s earlier findings on the extent to which members influence overall trip patterns. They also highlight the presence of morning and evening peaks in system use. Casual user activity appears to be the most concentrated in Cambridge on weekdays, though casual user activity in downtown Boston increases on the weekends. Member trips comprise most of the trip activity in northern parts of Cambridge, in Somerville, and in Brookline. Travel near universities—including but not limited to Harvard, MIT, Northeastern, Tufts, and Boston University—comprises a considerable portion of trip activity during the midday and evening/night periods.

 

FIGURE 3-9
High Volume O-D Pairs
by User Type and Day-and-Time Period

 

FIGURE 3-9: High Volume O-D Pairs by User Type and Day-and-Time Period: This series of 18 maps shows Hubway trip origin-destination (O-D) pairs identified in Hubway trip data, categorized by trip volume. Each map reflects trips made by a particular user group (member, casual user or both) and day-and-time period. These origin-destination pairs are based on trips taken between April 17, 2015 and December 18, 2015.

 

The remaining three percent of trips during the high-activity season have a common origin and destination point. Of these, 57 percent were made by casual users and 43 percent were made by members. Table 3-5 shows the top stations in terms of overall trip volumes during the high-activity season.

 

TABLE 3-5
Stations with High Volumes of “One-Station” Trips


Station

Number of Trips (Common O-D)

The Esplanade

1,296

MIT: Mass Ave

912

Harvard Square: Mass Ave @ Dunster St

848

Old State House

768

Boston Public Library

684

Charles Circle

665

TD Garden

636

Charlestown: Warren St

634

Total Common Origin and Destination Trips

6,443

Data source: 2015 Hubway trip logs (4/17/2015–12/18/2015).

O-D = Origin and destination.

 

Many of these stations are in areas with recreational and tourist activity, suggesting that many trips made from these locations may be for recreation, rather than strictly for transportation from one point to another. Many of these trips also could be made by people who are not familiar with the area. These eight stations account for nearly 20 percent of all trips with a common origin and destination.

 

4          Hubway’s Potential to Complement or Compete with Transit

4.1      Comparing Hubway and Transit Trip Characteristics

Section 3 provided an overview of the Hubway system, and where, when, and how many riders used it in 2015. This section explores instances when trips on the Hubway system may complement activity on the transit network or when Hubway trips may be competing with adequate transit service. Existing research has examined the ways that Hubway bikeshare service, when integrated with the MBTA system, can reduce travel time and transfers for trips throughout a combined network.35 Chiraphadhanakul modeled optimal travel on all links on the MBTA system, both with Hubway service (as it existed in 2011 and 2012) and without it, and found that the availability of Hubway improved travel time savings and reduced transfers between transit nodes, or connection points, across the combined system.36 His research highlights the overall value of strategically placed one-way links to extend the transit system or close gaps in the network.

 

This analysis uses trip data to look more closely at where and when large volumes of Hubway trips are being made, and explores the MBTA fixed-route transit alternatives that are available for those high-volume O-D pairs. In some cases, Hubway may complement transit by:      

 

However, in other cases, people may be using Hubway as a substitute for trips that might be as direct and as fast—or faster—on transit, and on links where travel on transit vehicles may be comfortable (relatively uncrowded). In these cases, it might be worth gathering more data to learn why people might use Hubway for these O-D pairs. 

 

4.2    Tools for Comparing Travel Times

To learn more about how transit service may compare to Hubway service for various O-D pairs, CTPS used a trip planning tool to generate alternative transit or walk itineraries for 2015 Hubway bikesharing trips. The goal of this exercise was to model how riders might otherwise have traveled on transit or by foot to complete the trips they made using Hubway, so as to compare travel times and other trip characteristics across modes. Staff used a methodology initially implemented by James Wong to compare bicycling travel time to transit travel time for trips between Capital Bikeshare stations in the Washington DC region.37 Following his approach, CTPS used Open Trip Planner (OTP), an open-source journey-planning tool, to generate transit or walk itineraries for individual Hubway trips from the 2015 trip dataset. 38 

 

Output information for these itineraries included the following:

 

 

CTPS used the Hubway trip travel time from the data logs and the alternate transit trip travel time from OTP’s output to calculate a travel-time ratio using the following formula:

 

Travel Time Ratio = Transit Travel Time (excluding walk access and egress time) / Hubway Travel Time       

 

As mentioned in Section 3.1, the Hubway trip data used in this analysis only included the portion of each Hubway trip from when a rider picked up a bicycle at the origin Hubway station to when she dropped it off at the destination Hubway station. The rider’s true origin (for example, her home), her true destination (for example, her workplace), and the distance she traveled between these locations and Hubway stations are all unknown. The Hubway station coordinates were the only data that CTPS could use to identify the locations associated with a Hubway trip and to model an alternative transit trip. However, using the Hubway stations as trip start-and-end points systematically underestimates the total time of the Hubway trip. Using these stations as start-and-end points for an alternative transit trip also may overestimate or underestimate travel time by transit, depending on the locations of the rider’s true origin and destination. To mitigate this bias, and better compare Hubway trips and their estimated transit alternatives, CTPS excluded walk access and walk egress times from the estimates of transit travel time that were used to calculate trip travel-time ratios. For more information about this adjustment, see Appendix C.

 

At this stage, CTPS removed several other groups of trips from the 2015 sample before conducting further analysis. (See the diagram in Appendix A for impacts on the overall trip sample.) These trips include the following:

 

4.3    Results for Individual Trips

CTPS used the scheme described in Table 4-1 to classify the travel time ratios that were calculated using the formula cited in Section 4.2.

 

TABLE 4-1
Travel-Time-Ratio Categories

Category

Travel-Time-Ratio Range

Transit Faster than Hubway

0 < x < 0.9

Transit Comparable to Hubway

    •  ≤ x ≤ 1.1

Transit Slower than Hubway

1.1 < x ≤ 2

Transit Much Slower than Hubway

x > 2

Source: Central Transportation Planning Staff.

 

Figure 4-1 summarizes how the transit alternatives for approximately 814,200 Hubway member trips were distributed across the travel-time-ratio categories listed in Table 4-1.

 

FIGURE 4-1
Alternatives to Hubway Trips by Travel-Time-Ratio Category

FIGURE 4-1: Alternatives to Hubway Trips by Travel-Time-Ratio Category: This chart shows the distribution of Open Trip Planner (OTP) alternate itineraries that were generated for Hubway member trips. Trip alternatives were categorized by their travel-time-ratio category (for transit alternatives) or if OTP generated a “walk-only” alternative itinerary.

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/201512/18/2015 (trip sample = 814,225).

Note: The Walk category includes some trips without itineraries.

 

In only 11 percent of cases would Hubway trips have been faster or comparable via transit. A larger share of trips (18 percent) would have taken more than twice as long by transit.

 

Figure 4-2 summarizes the modes included in the alternative transit itineraries for these Hubway trips, which were recommended by OTP to minimize travel time. Figure 4-3 provides similar information but excludes trips with walk-only itineraries. 

 


 

FIGURE 4-2
Alternatives to Hubway Trips by Mode

FIGURE 4-2: Alternatives to Hubway Trips by Mode: This chart shows the distribution of Open Trip Planner (OTP) alternate itineraries that were generated for Hubway member trips. Trip alternatives were categorized by the modes OTP included in the alternate itinerary.

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015 (trip sample = 814,225).

Notes: 1) The “Rapid Transit Only” and “Bus Only” categories include trips with transit itineraries that would have involved transfers between rapid transit lines or bus routes, respectively. 2) The Walk category includes some trips without itineraries.

 

FIGURE 4-3
Transit Alternatives to Hubway Trips by Mode (Walk Trips Excluded)

FIGURE 4-3: Transit Alternatives to Hubway Trips by Mode (Walk Trips Excluded):  This chart shows the distribution of Open Trip Planner (OTP) alternative transit itineraries that were generated for Hubway member trips. Trip alternatives were categorized by the modes OTP included in the alternative itinerary.  If OTP generated a “walk-only” alternate itinerary for a trip, the trip was not included in this chart.

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015 (trip sample (excluding walk trips) = 562,831).

Note: The “Rapid Transit Only” and “Bus Only” categories include trips with transit itineraries that would have involved transfers between rapid transit lines or bus routes, respectively.

 

Figure 4-2 shows that OTP recommended a bus trip alternative for 36 percent of member Hubway trips. For 11 percent of trips, OTP recommended a transit alternative that included both bus and rapid transit; this was the largest group of trips that would have involved multiple transit modes. Figures 4-1 and 4-2 both show that walk-only alternatives were suggested for 31 percent of Hubway trips.43 Figure 4-3 shows that slightly more than half of trips for which OTP recommended a transit alternative only involved bus service.

 

The next series of graphs show how shares of Hubway member trips vary throughout the day with respect to travel-time-ratio categories and OTP-recommended alternative modes. Figure 4-4 provides some context by showing the distribution of Hubway member trips over the course of a weekday. 

 

 

FIGURE 4-4
Hubway Member Trips by Weekday Hour

FIGURE 4-4: Hubway Member Trips by Weekday Hour: This chart shows the variation in the number of Hubway member trips that take place over the course of a weekday. This chart reflects member trips taken between April 17, 2015 and December 18, 2015.

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015 (trip sample (weekday trips) = 612,917).

 

Figure 4-5 shows how member trips that took place during each weekday hour fall into the various travel-time-ratio categories.

 

FIGURE 4-5
Transit Alternatives to Hubway Trips
 by Travel-Time Ratio Category and Weekday Hour

FIGURE 4-5: Transit Alternatives to Hubway Trips by Travel-Time Ratio Category and Weekday Hour: This chart shows the variation in Open Trip Planner (OTP) alternate itineraries for Hubway member trips taken at various times over the course of a weekday. For Hubway member trips where OTP generated a transit alternative, the chart reflects the applicable travel-time-ratio category. For Hubway member trips where OTP generated a walk-only alternative, the chart shows these trips as “walk-only.”

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015 (trip sample (weekday trips) = 612,917).

Figure 4-5 shows that even though the level of trip activity varies over the course of the day, the share of trips that would be faster or comparable in travel time by transit stays relatively consistent, though this share is largest in the late night or early morning hours. Meanwhile, the share of Hubway trips for which OTP recommended a walk alternative decreases during the morning and afternoon peak periods. This may be because transit service is operating at higher frequencies, and may be a more readily available alternative for certain trips.

 

Figure 4-6 shows the share of trips by recommended alternate mode, by weekday hour.

 

FIGURE 4-6
Hubway Trips by Alternative
Recommended Modes by Weekday Hour

FIGURE 4-6: Hubway Trips by Alternative Recommended Modes by Weekday Hour: This chart shows the variation in Open Trip Planner (OTP) alternate itineraries for Hubway member trips taken at various times over the course of a weekday. For Hubway member trips where OTP generated a transit alternative, the chart reflects the applicable travel-transit modes. For Hubway member trips where OTP generated a walk-only alternative, the chart shows these trips as “walk-only.”

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015, (trip sample (weekday trips) = 612,917).

 Note: The “Rapid Transit Only” and “Bus Only” categories include trips with transit itineraries that would have involved transfers between rapid transit lines or bus routes, respectively.

 

The share of trips for which OTP recommended a “bus-only” alternative fluctuates more dramatically over the course of the day than shares of trips with “rapid transit-only” or “bus and rapid transit” alternatives. This may be because bus service frequencies change more dramatically than do rapid transit service frequencies.  

 

Figure 4-7 shows how trips for which OTP recommended a bus-only, rapid-transit-only, or bus-and-rapid-transit alternative fell into the various travel-time-ratio categories. 

 

FIGURE 4-7
Hubway Trips by Alternative Recommended
Modes and Travel-Time-Ratio Category

FIGURE 4-7: Hubway Trips by Alternative Recommended Modes and Travel-Time-Ratio Category: This chart categorizes Hubway member trips by the transit alternatives generated for them by Open Trip Planner (OTP). First, trips are organized by whether bus, rapid transit, or both bus and rapid transit were included in the trip’s alternate itinerary. Then, in each category, the chart shows the distribution of trips according to their travel-time-ratio category.

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015.

Notes: 1) Data reflect both weekdays and weekends. All trips = 814,225 trips. 2) Walk trips (31 percent of trips) or trips involving other modes or combinations of modes (two percent of trips) are not shown. 3) The “Rapid Transit Only” and “Bus Only” categories include trips with transit itineraries that would have involved transfers between rapid transit lines or bus routes, respectively.

 

Of trips that had a “rapid-transit-only” alternative, 11 percent would have been faster and 12 percent would have been comparable in travel time if made by transit. These percentages are higher than in the “bus only” or “bus-and-rapid transit” categories. 

 

Alternative transit itineraries from OTP included the number of transfers between transit vehicles. Of the trips with transit itineraries, which make up 69 percent of all trips discussed in this section, 69 percent did not include any transfers, and another 30 percent involved one transfer. Figure 4-8 shows the number of trips in each travel-time-ratio category that have zero, one, or more than one transfers.

 

FIGURE 4-8
Hubway Trips by Number of Transfers
and Travel-Time-Ratio Category

FIGURE 4-8: Hubway Trips by Number of Transfers and Travel-Time-Ratio Category: This chart categorizes Hubway member trips by the transit alternatives generated for them by Open Trip Planner (OTP). First, trips are organized by travel-time-ratio category. Then, the chart shows the distribution of trips by the number of transfers included in their alternate itineraries.

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015 (trip sample = 562,831 trips with transit itineraries).

Note: Values are rounded to the nearest whole percent; not all columns sum to 100 percent.

 

The itineraries for Hubway trips that would have been faster or comparable in travel time by transit almost always involved zero transfers or one transfer.

 

Figure 4-9 shows Hubway trips with transit itineraries by their mode and number of transfers. Figure 4-10 incorporates information about how the alternate transit itineraries for these trips fell into the various travel-time-ratio categories.

 


 

FIGURE 4-9
Trips with Transit Itineraries by Mode and
Number of Transfers

FIGURE 4-9: Trips with Transit Itineraries by Mode and Number of Transfers: This chart categorizes Hubway member trips with alternate transit itineraries generated by Open Trip Planner (OTP). Trips are organized by the transit modes and number of transfers included in the alternate transit itineraries.

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015 (trip sample = 549,811 trips (of 562,831 trips with transit itineraries).

Note: Trips in other modal categories (two percent of trips with transit itineraries) are not shown.

 

FIGURE 4-10
Trips with Transit Itineraries by Mode,
Number of Transfers, and Travel-Time-Ratio Category

FIGURE 4-10: Trips with Transit Itineraries by Mode, Number of Transfers, and Travel-Time-Ratio Category:  This chart categorizes Hubway member trips with alternate transit itineraries generated by Open Trip Planner (OTP). Trips are organized by the transit modes and number of transfers included in the alternate transit itineraries. Within each category, the chart shows the number of trips in each travel-time-ratio category.

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015 (trip sample = 549,811 trips (of 562,831 trips with transit itineraries).

Notes: 1) Trips in other modal categories (two percent of trips with transit itineraries) are not shown. 2) Total trips in each category have been rounded to the nearest thousand.

There is a slightly larger share of Hubway trips with bus-and-rapid-transit alternative itineraries involving one transfer that are faster or comparable by transit (11 percent) than Hubway trips with “bus only” or “rapid transit only” itineraries involving one transfer (6 percent and 7 percent). For these trips, transit may be more competitive in terms of travel time because the alternative itineraries effectively combine the frequency and speed of rapid transit with the geographic coverage of the bus system.

 

4.4    Results for Origin-Destination Pairs

Sections 4.4 through 4.8 discuss Hubway trips that have been clustered into O-D pairs. O-D pairs can be categorized and analyzed using the metrics that CTPS used to analyze individual Hubway trips in section 4.3. The significant difference is that an O-D pair is essentially a bundle of Hubway trips taking place between a unique set of points. Hubway trips within each O-D bundle were taken at various times of day or on different days of the week, and OTP may have recommended different transit alternatives depending on when the trip took place. As a result, the alternate itineraries for these trips may fall into a range of travel-time-ratio categories, or involve a variety of modes. However, by examining trips according to their origin and destination, it is possible to see patterns in a geospatial context.

 

As described in Section 4.2, CTPS used OTP to generate alternative transit itineraries for approximately 1,043,000 Hubway trips.44 Approximately 814,000 of these trips were made by Hubway members. These member trips can be classified into 16,696 O-D pairs. CTPS analyzed O-D pairs that included 200 or more trips to ensure sufficient samples for analysis, and 873 O-D pairs met this criterion.

 

CTPS focused specifically on O-D pairs with large shares of Hubway trips that 1) would have been faster or comparable in travel time by transit; 2) would have been much slower by transit; or 3) would have had walk-only itineraries. Tables 4-2, 4-3, and 4-4 categorize the 873 O-D pairs with respect to the shares of their alternative itineraries that fall into these categories.

 

Table 4-2 shows that for 90 percent of the analyzed O-D pairs, less than 25 percent of trips would have been faster or comparable in travel time by transit. These results correspond to those in Figure 4-1, which shows that of all Hubway trips—and not just O-D pairs—only 11 percent would have been faster or comparable in travel time by transit. 

TABLE 4-2
O-D Pairs by Percentage of Trips that
Would Have Been Faster or Comparable by Transit
 

Share of Faster
or Comparable Trips

Number of

O-D Pairs

Percent of

O-D Pairs

100%

0

0%

75% to 99%

13

1%

50% to 74%

28

3%

25% to 49%

55

6%

1% to 24%

442

51%

0%

335

38%

Total O-D Pairs

873

100%

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015.

Note: Table shows data for 873 pairs, each of which includes 200 or more trips.

 

Table 4-3 shows that for 77 percent of O-D pairs, less than 25 percent of trips would have been much slower—more than twice as long—by transit. Only 12 percent of O-D pairs had more than 50 percent of trips that would have been much slower by transit than they would have been by Hubway. 

 

TABLE 4-3
O-D Pairs by Percentage of Trips that
Would Have Been Much Slower by Transit
 

Share of Much Slower Trips

Number of

O-D Pairs

Percent of

O-D Pairs

100%

1

0%

75% to 99%

40

5%

50% to 74%

57

7%

25% to 49%

101

12%

1% to 24%

313

36%

0%

361

41%

Total O-D Pairs

873

100%

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015.

Note: Table shows data for 873 pairs, each of which includes 200 or more trips.

 

Table 4-4 shows that most O-D pairs either 1) only included trips that would have had walk-only itineraries; or 2) did not include any trips that would have had walk-only itineraries. In 30 percent of O-D pairs, all Hubway trips would have had walk-only itineraries. Only one-third of O-D pairs include a mix of trips that would have had walk-only itineraries and those that would have had transit itineraries.

 


 

TABLE 4-4
O-D Pairs by Percentage of Trips that
Would Have Had Walk-Only Itineraries

Share of Walk Only Trips

Number of

O-D Pairs

Percent of
O-D Pairs

100%

263

30%

75% to 99%

88

10%

50% to 74%

48

5%

25% to 49%

54

6%

1% to 24%

93

11%

0%

327

37%

Total O-D Pairs

873

100%

Data source: 2015 Hubway trip logs and Open Trip Planner output, 4/17/2015–12/18/2015.

Note Table shows data for 873 pairs, each of which includes 200 or more trips.

 

Table 4-5 displays details for O-D pairs that include more than 1,000 member trips. These pairs comprise approximately five percent of all trips in the analysis sample (approximately 814,000 member trips in total). Many of these O-D pairs have very large shares of trips that would have had walk-only itineraries or much slower transit itineraries, although a few O-D pairs include trips from a variety of categories. Figure 4-11 shows these pairs on a map of the Hubway service area.

 

TABLE 4-5
High Volume O-D Pairs with Trips Classified by Travel-Time-Ratio Category

 

 

 

Percent of Total Trips

O-D Pair:
Station Numbers

O-D Pair: Station Names

Total Number of Trips

Faster by Transit

Comparable by Transit

 Slower by Transit

 Much Slower by Transit

Walk-
Only

Grand Total

M32025 to S32006

Linear Park: Mass Ave @ Cameron Ave to Davis Square

2,800

0%

0%

0%

0%

100%

100%

S32006 to M32025

Davis Square to Linear Park: Mass Ave @ Cameron Ave

2,542

0%

0%

0%

0%

100%

100%

A32010 to D32022

South Station to TD Garden

2,445

2%

2%

64%

32%

0%

100%

S32011 to S32006

Teele Square to Davis Square

1,962

1%

1%

1%

0%

96%

100%

D32022 to A32010

TD Garden to South Station

1,876

4%

10%

73%

14%

0%

100%

D32022 to C32010

TD Garden to Congress St @ Sleeper St

1,826

1%

1%

68%

30%

0%

100%

D32006 to A32010

Lewis Wharf to South Station

1,792

4%

5%

21%

5%

65%

100%

M32006 to B32016

MIT: Mass Ave to Beacon St @ Mass Ave

1,758

3%

4%

33%

6%

53%

100%

C32010 to D32022

Congress St @ Sleeper St to TD Garden

1,689

0%

1%

23%

76%

0%

100%

D32008 to D32022

Rowes Wharf to TD Garden

1,656

0%

0%

4%

3%

93%

100%

B32016 to M32006

Beacon St @ Mass Ave to MIT: Mass Ave

1,597

4%

10%

38%

2%

46%

100%

A32010 to D32006

South Station to Lewis Wharf

1,580

0%

2%

28%

4%

65%

100%

D32012 to D32022

Post Office Square to TD Garden

1,545

1%

2%

53%

26%

18%

100%

B32004 to A32010

Aquarium Station to South Station

1,340

0%

0%

4%

0%

96%

100%

M32006 to M32011

MIT: Mass Ave  to Central Square

1,315

2%

4%

57%

15%

23%

100%

A32010 to B32004

South Station to Aquarium Station

1,266

0%

0%

5%

0%

95%

100%

D32021 to D32022

Charlestown: Warren St to TD Garden

1,246

0%

0%

0%

0%

100%

100%

D32023 to D32022

Spaulding Hospital: Charlestown Navy Yard to TD Garden

1,221

2%

3%

67%

23%

5%

100%

D32022 to A32013

TD Garden to John F Fitzgerald - Surface Rd

1,210

0%

0%

0%

0%

100%

100%

M32011 to M32006

Central Square to MIT: Mass Ave

1,175

8%

17%

64%

1%

10%

100%

D32022 to D32023

TD Garden to Spaulding Hospital: Charlestown Navy Yard

1,175

7%

8%

54%

25%

6%

100%

A32010 to D32008

South Station to Rowes Wharf

1,109

0%

0%

0%

0%

100%

100%

M32003 to M32019

One Broadway / Kendall Square to CambridgeSide Galleria

1,079

0%

0%

0%

0%

100%

100%

S32006 to S32011

Davis Square to Teele Square

1,052

0%

1%

5%

0%

94%

100%

D32022 to D32021

TD Garden to Charlestown: Warren St

1,017

0%

0%

0%

0%

100%

100%

M32011 to M32014

Central Square to Harvard Housing: Putnam Ave

1,003

0%

0%

2%

1%

97%

100%

Data source: 2015 Hubway Trip Logs and OTP output, 4/17/2015–12/18/2015.

O-D = Origin-Destination. OTP = Open Trip Planner.

Note: Hubway trips within each O-D bundle were taken at various times of day or on different days of the week, and OTP may have recommended different transit alternatives depending on when the trip took place. As a result, the alternate itineraries for these trips may fall into a range of travel-time-ratio categories, or involve a variety of modes.

 

FIGURE 4-11
O-D Pairs with 1,000 or More Hubway Trips by Primary Travel-Time-Ratio Category


FIGURE 4-11: O-D Pairs with 1,000 or More Hubway Trips by Primary Travel-Time-Ratio Category: This map shows origin-destination (O-D) pairs of Hubway member trips. These O-D pairs are classified according to their trip volume and the dominant travel-time-ratio category (including the walk-only category) for trips in the pair, based on alternate itineraries generated by Open Trip Planner (OTP).

4.5      Identifying O-D Pairs that May Complement or Compete with Transit

To understand how Hubway O-D pairs may complement or compete with transit service, CTPS began by examining how the alternate transit itineraries for the trips in each O-D pair were distributed across the travel-time-ratio categories.

 

  1. If an O-D pair includes a large share of Hubway trips that would have been faster or comparable in travel time by transit, it is possible that Hubway use between that O-D pair may compete with the transit system, particularly if these trips would be direct and not crowded. However, Hubway service may play a complementary role
    1. if transit alternatives for these trips would involve transfers; or
    2. if Hubway trips were taking place during peak periods, and the transit vehicles serving them likely would be uncomfortable because of high levels of passenger demand. 45

Section 4.6 describes CTPS’s analysis of these O-D pairs.

 

  1. If most or all trips for a certain O-D pair would be much slower by transit, Hubway service could complement transit by giving regular transit riders a quicker, more direct alternative, especially if it would not be cost-effective for the MBTA to improve these links in the transit network. However, should the MBTA make improvements that would make these trips quicker and more direct by transit, Hubway service on these links could become competitive. Section 4.7 discusses these cases.
  1. If OTP recommended that a person otherwise walk for all (or almost all) Hubway trips for a particular O-D pair, transit likely would not be a suitable mode for making this connection. This could occur because 1) the trip might be relatively short and/or 2) transit service might be limited, involve transfers, or be non-existent for this O-D pair. In these instances, Hubway may complement transit. Section 4.8 examines these cases.
  1. Hubway may complement the MBTA transit network in cases where OTP recommended that a walk-only alternative for all (or almost all) Hubway trips between a particular O-D pair and one of the Hubway stations is located near a transit station. Using Hubway between these O-D pairs could help people gain access to the transit network. Section 4.8 also examines these cases.

 

4.6    Results for O-D Pairs with Large Shares of Trips that are Faster or Comparable in Travel Time by Transit

CTPS first examined Hubway trips and O-D pairs using the day-and-time periods shown in Table 4-6.

 

TABLE 4-6
Day-and-Time Periods

Time Period

Relevant Hours

Weekday - Late Night / Early Morning

12:00 AM–6:59 AM

Weekday AM Peak

7:00 AM–10:00 AM

Weekday - Midday

10:01 AM–3:59 PM

Weekday - PM Peak

4:00 PM–7:00 PM

Weekday - Evening / Night

7:01 PM–11:59 AM

Weekend

All day

Source: Central Transportation Planning Staff.

 

For each day-and-time period, CTPS selected a subset of pairs that met the following two criteria:

 

Trips between these O-D pairs may compete with transit during off-peak periods. They may also compete with transit during peak periods unless there is evidence of uncomfortable conditions or a need to make transfers on the transit modes that might otherwise serve the trip.

 

As discussed in Sections 4.2 and 4.3, the OTP output data provides information on the modes and transfers included in each transit itinerary. Using this information, CTPS reviewed the O-D pairs that met the trip volume and travel-time-ratio criteria during the different day-and-time periods and identified the combinations of transit modes and transfers that would have provided alternatives to Hubway trips for each O-D pair. CTPS documented the transit mode and transfer combinations that accounted for the majority of Hubway trips that would have been faster or comparable in travel time by transit. For more information about the results for these O-D pairs, see tables E-1 to E-4 in Appendix E.

 

CTPS considered Hubway activity on prominent O-D pairs during off-peak periods to be competing with transit. In general, the transit alternatives for these trips did not involve any transfers (for more information, see tables E-3 and E-4 in Appendix E). For O-D pairs that were prominent during the AM and PM peak periods, CTPS followed several additional steps:

 

 

Figure 4-12 shows the results of this four-step analysis for Hubway trips that took place in the AM and PM peak periods. The O-D pairs shown in Figure 4-12 are classified by the transit mode that served the majority of trips that would be faster or comparable in travel time by transit. Tables E-1 and E-2 in Appendix E contain more detailed information for the OD pairs in these time periods.

 

FIGURE 4-12
O-D Pairs that may Complement or Compete
with Transit (AM and PM Peak Period)

FIGURE 4-12: O-D Pairs that May Complement or Compete with Transit (AM and PM Peak Periods): This series of two maps shows origin-destination (O-D) pairs of Hubway member trips. One map shows O-D pairs during the weekday AM peak period, and the other shows O-D pairs during the weekday PM peak period. These O-D pairs are classified according to their trip volume, the relevant modes in the alternate transit itineraries generated by Open Trip Planner (OTP), and whether the trips may complement or compete with transit. At least 50 percent of the trips in these pairs were faster or comparable in travel time by transit. More information about these O-D pairs is available in Appendix E.



 

Many of the transit alternatives for the O-D pairs in the weekday AM peak period involve rapid transit, while a wider variety of transit modes could serve the pairs in the weekday PM peak period. When these two periods are compared, it is possible to see how the flow of travel changes over the course of the day. Three pairs that flow away from TD Garden during the AM peak period flow toward TD Garden during the PM peak period.

 

Most O-D pairs in both periods could be complements to transit, because some evidence suggests that transit travel on these segments could be uncomfortable during peak periods. Transit alternatives for these O-D pairs did not involve transfers. In both periods, many alternative transit trips would have relied upon the Green Line subway and the portion of the Orange Line between Back Bay and North Station. Commuter rail would have supported transit trips between several O-D pairs that may compete with transit; there is no evidence of uncomfortable conditions for these commuter rail segments. However, commuter rail is not necessarily designed to help people make these short-distance connections in urban cores. In particular, commuter rail service is less frequent than other modes, such as rapid transit; as a result, riders need to be aware of schedules, and their trips would need to coincide with these schedules. People might also avoid commuter rail for this O-D pair because they might assume that it is more expensive and less convenient to travel by commuter rail, even if the trip would be quick and presumably comfortable.

 

Figure 4-13 shows O-D pairs that met the trip volume and travel-time-ratio criteria during the weekday off-peak periods and the weekend period. Tables E-3 and E-4 in Appendix E contain more detailed information for the OD pairs in these time periods, including the shares of faster or comparable trips that could be served by different transit modes. The O-D pairs shown in Figure 4-13 are classified by the transit mode that served the majority of these trips. CTPS did not investigate whether conditions would be uncomfortable on relevant transit modes during these off-peak periods; these O-D pairs could be considered potential competitors to transit until more detailed transit itinerary and passenger demand information is analyzed. The maps in Figure 4-13 provide information about the transit mode combinations that would otherwise serve these O-D pairs. Figure 4-13 shows that Hubway O-D pairs near Harvard, MIT, Tufts, and Boston University are prominent during the weekday midday and evening periods and during the weekend. During the early-morning and midday periods, several O-D pairs that have many trips with transit alternatives that would be faster or comparable in travel time to Hubway would be served by rapid transit. In the evening and on the weekend, more of these O-D pairs would be served by buses.

 

FIGURE 4-13
O-D Pairs that may Compete
with Transit (Weekday Off-Peak and Weekend Periods)

 

FIGURE 4-13: O-D Pairs that May Complement or Compete with Transit (Weekday Off-Peak and Weekend Periods): This series of four maps shows origin-destination (O-D) pairs of Hubway member trips. The first shows O-D pairs during the weekday late night/early morning period, the second shows O-D pairs during the weekday midday period, the third shows O-D pairs during the weekday evening/night period, and the fourth shows O-D pairs during weekend days. . These O-D pairs are classified according to their trip volume, the relevant modes in the alternate transit itineraries generated by Open Trip Planner (OTP). At least 50 percent of the trips in these pairs were faster or comparable in travel time by transit. More information about these O-D pairs is available in Appendix E.

4.7    Results for O-D Pairs with Large Shares of Trips that are Much Slower by Transit

As shown in Table 4-1, Hubway trips that would have been much slower by transit are those for which OTP recommended an alternative transit itinerary with a travel time that was more than twice as long as the Hubway travel time. Given this difference in travel time, O-D pairs that include a large share of these trips may complement transit because it might be difficult for the MBTA to provide a quick or direct transit option for this link. This is particularly the case if Hubway riders also have MBTA monthly passes, which would mean that the MBTA would receive the same revenues even if these people were using Hubway to make these trips. However, if the MBTA can apply a relatively low-cost fix to improve transit service that would otherwise serve these links, Hubway could eventually compete with transit. In either case, it is important to understand where O-D pairs with large volumes of “much slower” trips are located, so that the MBTA can choose whether to improve transit service in these locations.  

 

To identify relevant O-D pairs for analysis, CTPS examined each day-and-time period listed in Table 4-3 and selected a subset of O-D pairs that met the following two criteria:

 

Figures 4-14 and 4-15 show these O-D pairs by time of day. These individual O-D pairs have been classified by the transit mode that would have served 50 percent or more of trips that would have been much slower by transit. Tables F-1 to F-5 in Appendix F describe these O-D pairs. These tables show that for each O-D pair, one transit mode-and-transfer combination tends to account for most trips that would have been slower or much slower by transit. For example, the alternative option for 85 percent of “much slower” trips between Inman Square and the MIT Stata Center during the AM peak period would be a bus trip with no transfers. However, although one option tends to dominate, the transit options for “much slower” trips on many O-D pairs are spread across multiple mode-and-transfer categories.

 

Figure 4-14 shows that during the weekday AM peak period, many O-D pairs meet the trip volume and travel-time-ratio criteria, though no pair stands out in terms of trip volume. In the PM peak period, there are fewer O-D pairs that meet these criteria, but the qualifying O-D pairs vary more by volume. The highest volume O-D pairs generally connect TD Garden (North Station) to points in Cambridge or South Boston. During the PM peak period, several O-D pairs represent trips flowing in the opposite direction of O-D pairs during the AM peak period, although different combinations of transit modes might have served these trips during the PM peak period.

                                                            

Figures 4-14 and 4-15 both show that OTP recommended bus options for many O-D pairs that would have been much slower by transit. Tables F-1 to F-5 in Appendix F show that transit itineraries for these OD pairs involve more transfers and multi-modal combinations than do the itineraries for O-D pairs with 50 percent or more trips that would be faster or comparable by transit.

 

FIGURE 4-14
O-D Pairs with 75 Percent or More Trips Much Slower by Transit
(Weekday AM and PM Peak Periods)

FIGURE 4-14: O-D Pairs with 75 Percent or More Trips Much Slower by Transit (Weekday AM and PM Peak Periods): This series of two maps shows origin-destination (O-D) pairs of Hubway member trips. One map shows O-D pairs during the weekday AM peak period, and the other shows O-D pairs during the weekday PM peak period. These O-D pairs are classified according to their trip volume and the relevant modes in the alternate transit itineraries generated by Open Trip Planner (OTP). At least 75 percent of the trips in these pairs were faster or comparable in travel time by transit. More information about these O-D pairs is available in Appendix F.

FIGURE 4-15
O-D Pairs with 75 Percent or More Trips Much Slower by Transit
(Weekday Off- Peak and Weekend Periods)
FIGURE 4-15: O-D Pairs with 75 Percent or More Trips Much Slower by Transit (Weekday Off-Peak and Weekend Periods): This series of three maps shows origin-destination (O-D) pairs of Hubway member trips. The first shows O-D pairs during the weekday late night/early morning period, the second shows O-D pairs during the weekday midday period, and the third shows O-D pairs during weekend days. . These O-D pairs are classified according to their trip volume, the relevant modes in the alternate transit itineraries generated by Open Trip Planner (OTP). At least 75 percent of the trips in these pairs were faster or comparable in travel time by transit. More information about these O-D pairs is available in Appendix F.


4.8    Results for O-D Pairs with Large Shares of Walk-Only Trips

For some connections, Hubway may complement the existing transit system by providing a way for people to make trips on links where transit service is limited, complicated, nonexistent, or otherwise not a sensible alternative (especially if the distance between the origin and destination is relatively short). CTPS used OTP outputs that recommended a walk-only alternative (as opposed to a transit alternative) for many trips between a particular O-D pair as a way to identify these cases. As part of this analysis, CTPS examined O-D pairs during each day-and-time period listed in Table 4-3 and identified a subset of O-D pairs that met the following two criteria:

 

CTPS also classified Hubway stations according to the transit modes available near each station,  using a geographic information systems (GIS) analysis of the bus stops and modes available at transit stations within 200 meters (approximately one-eighth mile) of each Hubway station. Figure 4-16 classifies Hubway stations by the transit modes that are nearby. Appendix B also lists the transit modes that are near each Hubway station.

 

Figure 4-17 shows the high-volume O-D pairs with 95 percent or more walk-only trips during the weekday AM and PM peak periods and Figure 4-18 shows these pairs for the weekday off-peak and weekend periods. Tables G-1 to G-6 in Appendix G list the prominent O-D pairs during each time period.

 

As Figures 4-17 and 4-18 show, many of these O-D pairs exist along the Rose Kennedy Greenway in Boston, near MIT and Harvard, and near stations further north on the Red Line, including Porter and Davis Squares.


FIGURE 4-16
Hubway Stations by Transit Modes Available within 200 Meters

FIGURE 4-16: Hubway Stations by Transit Modes Available within 200 Meters. This map classifies Hubway stations by the transit modes that are available within 200 meters of the Hubway station. More information on individual stations and the transit modes near to them is available in Appendix B.

 

FIGURE 4-17
O-D Pairs with 95 Percent or More Walk-Only Trips
(Weekday AM and PM Peak Periods)

FIGURE 4-17: O-D Pairs with 95 Percent or More Walk-Only Trips (Weekday AM and PM Peak Periods): This series of two maps shows origin-destination (O-D) pairs of Hubway member trips. One map shows O-D pairs during the weekday AM peak period, and the other shows O-D pairs during the weekday PM peak period. These O-D pairs are classified according to their trip volume. At least 95 percent of the trips in these pairs had “walk-only” travel itineraries generated by Open Trip Planner (OTP). More information about these O-D pairs is available in Appendix G. The maps also classify Hubway stations by the transit modes that are accessible within 200 meters.

FIGURE 4-18
O-D Pairs with 95 Percent or More Walk-Only Trips(Weekday Off-Peak and Weekend Periods)

FIGURE 4-18: O-D Pairs with 95 Percent or More Walk-Only Trips (Weekday Off-Peak and Weekend Periods):  This series of four maps shows origin-destination (O-D) pairs of Hubway member trips. The first shows O-D pairs during the weekday late night/early morning period, the second shows O-D pairs during the weekday midday period, the third shows O-D pairs during the weekday evening/night period, and the fourth shows O-D pairs during weekend days. These O-D pairs are classified according to their trip volume. At least 95 percent of the trips in these pairs had “walk-only” travel itineraries generated by Open Trip Planner (OTP). More information about these O-D pairs is available in Appendix G. The maps also classify Hubway stations by the transit modes that are accessible within 200 meters.

CTPS analyzed which O-D pairs with large shares of walk-only trips might complement transit by helping riders connect to stations, specifically those serving rapid transit. As shown in Figure 2-21, more than half of Hubway member survey respondents reported making at least one trip that ended near a rapid transit station during a typical week. To identify this subset of O-D pairs, CTPS examined the transit modes that were near the origin and destination stations of each O-D pair. Figures 4-19 and 4-20 take the O-D pairs shown in Figures 4-17 and 4-18 and classify them by whether they include zero, one, or two trip ends at a rapid transit station.

 

FIGURE 4-19
O-D Pairs with 95 Percent or More Walk-Only Trips by Connections to Rapid Transit
(Weekday AM and PM Peak Periods)

FIGURE 4-19: O-D Pairs with 95 Percent or More Walk-Only Trips by Connections to Rapid Transit (Weekday AM and PM Peak Periods): This series of two maps shows origin-destination (O-D) pairs of Hubway member trips. One map shows O-D pairs during the weekday AM peak period, and the other shows O-D pairs during the weekday PM peak period. These O-D pairs are classified according to their trip volume and the number of trip ends that were near transit, particularly rapid transit. At least 95 percent of the trips in these pairs had “walk-only” travel itineraries generated by Open Trip Planner (OTP). More information about these O-D pairs is available in Appendix G. The maps also classify Hubway stations by whether or not transit, particularly rapid transit, is accessible within 200 meters.

 

FIGURE 4-20
O-D Pairs with 95 Percent or More Walk-Only Trips by Connections toRapid Transit
(Weekday Off-Peak and Weekend Periods)

FIGURE 4-20: O-D Pairs with 95 Percent or More Walk-Only Trips by Connections to Rapid Transit (Weekday Off-Peak and Weekend Periods):  This series of four maps shows origin-destination (O-D) pairs of Hubway member trips. The first shows O-D pairs during the weekday late night/early morning period, the second shows O-D pairs during the weekday midday period, the third shows O-D pairs during the weekday evening/night period, and the fourth shows O-D pairs during weekend days. These O-D pairs are classified according to their trip volume and the number of trip ends that were near transit, particularly rapid transit. At least 95 percent of the trips in these pairs had “walk-only” travel itineraries generated by Open Trip Planner (OTP). More information about these O-D pairs is available in Appendix G. The maps also classify Hubway stations by whether or not transit, particularly rapid transit, is accessible within 200 meters.
Figure 4-21 shows the number of O-D pairs in each day-or-time period that have zero, one, or both ends near a rapid transit station. High-volume O-D pairs that 1) are almost exclusively made up of walk alternative trips, and 2) have one trip end at a rapid transit station may support first-or-last-mile connections to rapid transit. Trips between O-D pairs with no ends near rapid transit may close gaps in the transportation network by providing service where transit service does not exist and/or where it may not make sense to provide fixed-route transit service. More analysis of trips between origins and destinations that each are near rapid transit would be needed to determine whether these trips might complement or compete with transit. If the origin and destination Hubway stations each are near different rapid transit lines, Hubway trips between these points could help travelers avoid transit trips with transfers.

 

FIGURE 4-21
O-D Pairs with 95 Percent or More Walk-Only Trips
 by Number of Ends within 200 Meters of Rapid Transit Stations

FIGURE 4-21: O-D Pairs with 95 Percent or More Walk-Only Trips by Number of Ends within 200 Meters of Rapid Transit Stations: This chart categorizes origin-destination (O-D) pairs by day-and-time period and the number of trip ends that are within 200 meters of rapid transit.

Data sources: 2015 Hubway trip logs (4/17/2015–12/18/2015), OTP output, CTPS GIS analysis.

Note: O-D pairs may appear in multiple day/time periods.


 

It is important to note that OTP’s algorithm determined whether to recommend a walk or transit alternative for a particular O-D pair using MBTA GTFS files that had been incorporated into the OTP implementation. This implementation does not include data for other transit services that may be operating in the Hubway service area—including Transportation Management Area, university, or other shuttles. If available data from these other services had been incorporated into OTP, it is possible that some of these O-D pairs would have had lower shares of walk alternative trips. That said, the fact that people are using Hubway to make these connections raises questions about whether people are aware of these other transit services or perceive them as being a preferable alternative to Hubway.

 

4.9    Summary of Findings

Section 3 describes when and where Hubway riders are making their trips, in terms of visits to stations or between O-D pairs. This information creates a context for the analysis in Section 4, which examines potential synergies and areas of competition between the Hubway system and the MBTA fixed-route transit system. CTPS used the ratio of Hubway travel time to the travel time from OTP’s alternative itinerary for each trip as a basis for comparing the two modes in different circumstances.

 

Some of these synergies and potential areas of competition between Hubway and transit appear at the trip level (see Section 4.3).

 

 

 

Sections 4.4 through 4.8 described how CTPS clustered trips into O-D pairs to explore trip flows based on time and geographic location. CTPS examined O-D pairs with larger-than-typical shares of trips with walk-only alternatives or transit alternatives that would have been faster, comparable to, slower, or much slower than the Hubway trips. The findings discussed in these sections generally highlight potentially complementary relationships between Hubway and MBTA fixed-route transit because Hubway might provide an alternative for

 

However, Hubway O-D pairs with large shares of trips in the slower range (travel time ratio between 1 and 2) or much slower range (travel time ratio greater than 2), merit further exploration. Hubway trips that would have much slower transit alternatives may be able to complement transit if it would not make sense for the MBTA to improve transit service or provide other options on these links. More nuanced analyses would be needed to identify the potential for synergies or competition for O-D pairs where OTP generally identified slower or much slower transit alternatives.

 

This study provides one analytical perspective on how bike sharing and the fixed-route transit system interrelate by using evidence of when and where people have used the Hubway system to meet their travel needs. However, Hubway trip data by itself cannot provide complete information about how and why people chose to use Hubway instead of transit or other modes in specific circumstances. This study is limited in that it is inherently biased toward Hubway by only analyzing Hubway trips, which likely happen in places where Hubway is convenient and transit is not. The data do not make it possible to analyze cases when people might have chosen to use transit instead of Hubway. Moreover, CTPS’s ability to identify possible alternatives to Hubway trips was limited by having data only on the portions of trips between the Hubway stations, as discussed in Section 4.2. Without knowing the true origins and destinations of riders’ full trips, or the purposes of their trips, it is not possible to identify an optimal itinerary definitively.

 

Future research could include a more detailed examination of the O-D pairs in Sections 4.6 to 4.8, in order to determine whether Hubway use on these links should be encouraged, or if transit service in these areas should be improved to make it more competitive. Other options include exploring the travel patterns of casual bikesharing users, including their potential to interact with other parts of the transportation system, particularly fixed-route transit. The methodology used in this report also could be adapted to compare the travel times of other point-to-point shared-use mobility services to transit.  


 

 

Appendix A:
Data Used in Analysis Stages

 


 

FIGURE A-1: Hubway Trips Used in Analysis Steps: This figure shows the sequence of steps that CTPS followed to analyze the Hubway 2015 trip state, and the number of trips that were included or removed at each step.

Appendix B:
Station Names, Aliases, and Visit Data

TABLE B-1
Hubway Station Details

 

 

 

 

 

 

 

 

4/17/2015 to 12/18/2015

 

Hubway Station Number

Original
Hubway Station
Name

Hubway
Station Alias

Trip Origins from Station

Trip Destinations at Station

Total Station Visits

Transit Options within 200 Meters

Boston Stations:

 

 

D32022

TD Garden - Causeway @ Portal Park #1 a

TD Garden

30,117

34,953

65,070

Bus, Commuter Rail, and Rapid Transit

A32010

South Station - 700 Atlantic Ave.

South Station

30,434

30,938

61,372

All

D32005

Boston Public Library - 700 Boylston St.

Boston Public Library

18,944

19,445

38,389

Bus and Rapid Transit

D32016

Charles Circle - Charles St. @ Cambridge St.

Charles Circle

17,285

16,823

34,108

Rapid Transit Only

C32003

Back Bay / South End Station

Back Bay / South End

16,473

15,637

32,110

Bus, Commuter Rail, and Rapid Transit

B32010

Kenmore Sq / Comm Ave

Kenmore Square

14,330

14,461

28,791

Bus and Rapid Transit

D32010

Cross St. @ Hanover St.

Cross St @ Hanover St

14,497

13,551

28,048

Bus and Rapid Transit

B32008

Old State House b

Old State House

14,217

13,364

27,581

Bus and Rapid Transit

B32016

Beacon St / Mass Ave

Beacon St @ Mass Ave

13,861

13,345

27,206

Bus Only

D32007

Boylston St. @ Arlington St.

Boylston St @ Arlington St

13,119

14,044

27,163

Bus and Rapid Transit

D32017

The Esplanade - Beacon St. @ Arlington St.

The Esplanade

13,204

12,914

26,118

None

C32008

Boylston @ Fairfield

Boylston St @ Fairfield St

12,646

12,664

25,310

Bus Only

D32000

Cambridge St. @ Joy St.

Cambridge St @ Joy St

12,826

12,436

25,262

Bus Only

C32007

Prudential Center / Belvidere

Prudential Center

12,404

12,697

25,101

Bus and Rapid Transit

D32008

Rowes Wharf - Atlantic Ave

Rowes Wharf

12,452

12,376

24,828

Bus Only

D32014

Tremont St / West St

Tremont St @ West St

12,226

12,218

24,444

Bus, Rapid Transit, and Silver Line

B32004

Aquarium Station - 200 Atlantic Ave.

Aquarium Station

12,206

11,733

23,939

Bus and Rapid Transit

B32005

Christian Science Plaza

Christian Science Plaza

11,729

11,500

23,229

Bus and Rapid Transit

B32018

Boylston / Mass Ave

Boylston St @ Mass Ave

11,328

11,235

22,563

Bus and Rapid Transit

D32012

Post Office Square

Post Office Square

10,841

11,171

22,012

Bus Only

D32006

Lewis Wharf - Atlantic Ave.

Lewis Wharf

11,101

10,663

21,764

Bus Only

D32021

Charlestown - Warren St @ Chelsea St

Charlestown: Warren St

11,227

10,531

21,758

Bus Only

B32000

Newbury St / Hereford St

Newbury St @ Hereford St

10,396

10,220

20,616

Bus and Rapid Transit

D32019

Boylston St / Washington St

Boylston St @ Washington St

10,165

10,342

20,507

Bus, Rapid Transit, and Silver Line

D32004

Franklin St. / Arch St.

Franklin St @ Arch St

10,301

10,176

20,477

Bus and Rapid Transit

C32010

Congress / Sleeper

Congress St @ Sleeper St

9,617

10,732

20,349

Bus Only

B32011

Yawkey Way @ Boylston St.

Yawkey Way @ Boylston St

9,786

9,760

19,546

Bus Only

C32006

Washington St. @ Rutland St.

Washington St @ Rutland St

9,373

9,592

18,965

Bus and Silver Line

B32003

HMS / HSPH - Ave. Louis Pasteur @ Longwood Ave.

Harvard: HMS/HSPH

9,425

9,279

18,704

Bus Only

D32013

Boylston St / Berkeley St

Boylston St @ Berkeley St

8,866

9,388

18,254

Bus Only

C32000

Tremont St. @ Berkeley St.

Tremont St @ Berkeley St

9,395

8,857

18,252

Bus Only

D32024

Charles St @ Beacon St

Charles St @ Beacon St

9,353

8,863

18,216

Bus Only

D32002

Washington St. @ Waltham St.

Washington St @ Waltham St

9,103

9,077

18,180

Bus and Silver Line

C32002

Boston Medical Center -  East Concord @ Harrison Ave

Boston Medical Center

8,913

9,037

17,950

Bus Only

A32002

Agganis Arena - 925 Comm Ave.

Agganis Arena

8,423

8,688

17,111

Bus and Rapid Transit

D32015

Chinatown Gate Plaza - Surface Rd. @ Beach St.

Chinatown Gate Plaza

8,240

8,717

16,957

Bus Only

A32004

Longwood Ave / Binney St

Longwood Ave @ Binney St

7,783

8,798

16,581

Bus Only

C32004

Columbus Ave. @ Mass. Ave.

Columbus Ave @ Mass Ave

8,551

7,688

16,239

Bus and Rapid Transit

B32015

Landmark Centre

Landmark Center

7,662

7,870

15,532

Bus Only

D32011

Stuart St. @ Charles St.

Stuart St @ Charles St

7,606

7,708

15,314

Bus Only

A32012

Packard's Corner - Comm. Ave. @ Brighton Ave.

Packard's Corner

7,730

7,512

15,242

Bus and Rapid Transit

A32009

Tremont St / W Newton St

Tremont St @ W Newton St

7,950

7,177

15,127

Bus Only

A32003

B.U. Central - 725 Comm. Ave.

BU Central

7,172

7,544

14,716

Bus and Rapid Transit

B32014

Seaport Hotel

Seaport Hotel

7,345

7,047

14,392

Bus and Silver Line

A32013

John F Fitzgerald - Surface Road @ India Street

John F Fitzgerald - Surface Rd

7,021

7,263

14,284

Bus and Rapid Transit

D32009

Faneuil Hall - Union St. @ North St.

Faneuil Hall

7,257

7,016

14,273

Bus and Rapid Transit

A32008 Buswell St. @ Park Dr. Buswell St @ Park Dr 6,777 7,028 13,805 Bus and Rapid Transit

A32006

Harvard University Housing - 111 Western Ave. @ Soldiers Field Park

Harvard Housing: Western Ave

6,502

6,308

12,810

None

B32002

Ruggles Station / Columbus Ave.

Ruggles Station

6,313

6,397

12,710

Bus, Commuter Rail, and Rapid Transit

B32007

Seaport Square - Seaport Blvd. @ Boston Wharf

Seaport Square

6,528

6,112

12,640

Bus Only

B32012

Northeastern U / North Parking Lot

Northeastern

6,121

6,239

12,360

Bus and Rapid Transit

D32027

New Balance Store - Boylston @ Dartmouth

New Balance: Boylston St

5,764

5,849

11,613

Bus and Rapid Transit

D32020

Charlestown - Main St @ Austin St

Charlestown: Main St

5,954

5,500

11,454

Bus Only

C32025

Ink Block

Ink Block

5,677

5,736

11,413

Bus and Silver Line

B32013

Brigham Cir / Huntington Ave

Brigham Circle

5,291

5,118

10,409

Bus and Rapid Transit

C32024

State Street @ Channel Center

State Street

5,133

5,178

10,311

None

B32006

Colleges of the Fenway

Colleges of the Fenway

5,017

5,004

10,021

Bus Only

A32019

Harvard University Transportation Services - 175 North Harvard St

Harvard: Transportation Services

5,066

4,843

9,909

Bus Only

D32023

Spaulding Rehabilitation Hospital - Charlestown Navy Yard

Spaulding Hospital: Charlestown Navy Yard

4,894

5,009

9,903

None

A32000

Fan Pier

Fan Pier

4,668

5,231

9,899

None

A32017

Allston Green District - Commonwealth Ave & Griggs St

Allston Green District

5,123

4,529

9,652

Rapid Transit Only

D32018

Boston Convention & Exhibition Center

BCEC

4,756

4,632

9,388

Bus and Silver Line

C32016

West Broadway @ Dorchester St

W Broadway @ Dorchester St

4,846

4,363

9,209

Bus Only

C32017

South Boston Library - 646 East Broadway

South Boston Library

4,352

4,562

8,914

Bus Only

C32001

Roxbury Crossing Station

Roxbury Crossing Station

4,244

4,015

8,259

Bus and Rapid Transit

C32021

ID Building West

ID Building W

3,844

3,957

7,801

Bus and Silver Line

A32001

Union Square - Brighton Ave. @ Cambridge St.

Union Square: Brighton Ave

3,662

3,723

7,385

Bus Only

C32009

Dorchester Ave. @ Gillette Park

Dorchester Ave @ Gillette Park

3,542

3,554

7,096

Bus and Rapid Transit

E32003

Hyde Square @ Barbara St

Hyde Square

3,458

3,131

6,589

Bus Only

C32020

ID Building East

ID Building E

3,221

3,341

6,562

Bus and Silver Line

E32005

Green St T

Green Street Station

3,133

3,270

6,403

Rapid Transit Only

A32011

Innovation Lab - 125 Western Ave. @ Batten Way

Innovation Lab: Western Ave

3,065

3,174

6,239

Bus Only

E32006

Jackson Square T @ Centre St

Jackson Square Station

3,142

2,932

6,074

Bus and Rapid Transit

B32020

BIDMC - Brookline @ Burlington St

BIDMC - Brookline

2,888

3,105

5,993

Bus Only

C32005

Washington St. @ Lenox St.

Washington St Lenox Street

2,748

2,826

5,574

Bus and Silver Line

C32012

Andrew Station - Dorchester Ave @ Humboldt Pl

Andrew Station

2,689

2,533

5,222

Bus and Rapid Transit

D32026

Hayes Square @ Vine St.

Hayes Square

2,445

2,612

5,057

Bus Only

E32002

JP Center - Centre Street @ Myrtle Street

JP Center

2,094

2,288

4,382

Bus Only

A32005

Harvard Real Estate - Brighton Mills - 370 Western Ave

Harvard Real Estate - Brighton Mills - 370 Western Ave

2,012

2,136

4,148

Bus Only

E32001

JP Monument - South St @ Centre St

JP Monument

2,053

2,059

4,112

Bus Only

B32017

Dudley Square

Dudley Square

1,601

1,602

3,203

Bus and Silver Line

C32013

JFK / UMASS Station

JFK / UMASS Station

1,431

1,610

3,041

Bus, Commuter Rail, and Rapid Transit

D32001

New Balance - 20 Guest St.

New Balance: Guest St

1,235

1,297

2,532

Bus Only

C32022

Newmarket Square

Newmarket Square

1,183

1,308

2,491

Commuter Rail, Bus, and Silver Line

E32004

Egleston Square @ Columbus Ave

Egleston Square

1,128

881

2,009

Bus Only

C32018

E. Cottage St @ Columbia Rd

E Cottage St @ Columbia Road

703

776

1,479

Bus Only

B32021

Wentworth Institute of Technology

Wentworth IT

711

705

1,416

Bus and Rapid Transit

C32019

Upham's Corner - Ramsey St @ Dudley St

Upham's Corner

761

609

1,370

Bus Only

C32029

West Broadway @ D Street

W Broadway @ D St

670

690

1,360

Bus Only

C32014

UMass Boston Integrated Sciences Complex

UMass Boston ISC

648

650

1,298

None

C32015

Mt Pleasant Ave / Dudley Town Common

Dudley Town Common

625

566

1,191

Bus Only

A32022

Main Street @ Eden Street Park

Main St @ Eden St Park

614

552

1,166

Bus Only

C32028

Lawn on D

Lawn on D

535

493

1,028

None

A32023

Bunker Hill Community College

Bunker Hill CC

252

226

478

Rapid Transit Only

D32028

Brighton Center

Brighton Center

202

197

399

Bus Only

E32007

Heath St @ South Huntington

Heath St @ S Huntington Ave

191

142

333

Bus and Rapid Transit

D32029

Washington St @ Brock St

Washington St @ Brock St

116

127

243

Bus Only

C32031

Ryan Playground - Dorchester Avenue Station

Ryan Playground

88

100

188

Bus Only

X32999

Dorrance Warehouse

Dorrance Warehouse

45

104

149

No Data

C32027

Franklin Park - Seaver Street @ Humbolt Ave

Franklin Park

65

39

104

Bus Only

C32030

Franklin Park Zoo

Franklin Park  Zoo

42

35

77

Bus Only

Brookline Stations:

 

 

K32001

Coolidge Corner - Beacon St @ Centre St

Coolidge Corner

6,213

6,197

12,410

Bus and Rapid Transit

K32003

Brookline Village - Station Street @ MBTA

Brookline Village

4,017

4,217

8,234

Bus and Rapid Transit

K32004

JFK Crossing @ Harvard St. / Thorndike St.

JFK Crossing

3,348

2,860

6,208

Bus Only

K32002

Washington Square @ Washington St. / Beacon St.

Washington Square

3,193

2,584

5,777

Bus and Rapid Transit

Cambridge Stations:

 

 

M32006

MIT @ Mass Ave / Amherst St

MIT: Mass Ave

26,511

26,630

53,141

Bus Only

M32018

Harvard Square @ Mass Ave/ Dunster

Harvard Square: Mass Ave @ Dunster St

20,856

21,538

42,394

Bus and Rapid Transit

M32011

Central Square @ Mass Ave / Essex St

Central Square

19,011

18,603

37,614

Bus and Rapid Transit

M32009

Lafayette Square @ Mass Ave / Main St / Columbia St

Lafayette Square

17,832

17,563

35,395

Bus Only

M32005

MIT Stata Center @ Vassar St / Main St

MIT: Stata Center

14,919

19,626

34,545

None

M32003

One Broadway / Kendall Sq @ Main St / 3rd St

One Broadway / Kendall Square

13,419

12,928

26,347

Bus and Rapid Transit

M32037

Ames St @ Main St

Ames St @ Main St

12,267

12,666

24,933

Bus and Rapid Transit

M32012

Central Sq Post Office / Cambridge City Hall @ Mass Ave / Pleasant St

Cambridge City Hall

12,414

11,964

24,378

Bus and Rapid Transit

M32002

One Kendall Square @ Hampshire St / Portland St

One Kendall Square

12,312

11,818

24,130

Bus Only

M32010

Inman Square @ Vellucci Plaza / Hampshire St

Inman Square

11,250

10,308

21,558

Bus Only

M32001

Lechmere Station @ Cambridge St / First St

Lechmere Station

9,969

10,087

20,056

Bus and Rapid Transit

M32022

Lower Cambridgeport @ Magazine St/Riverside Rd

Lower Cambridgeport

9,962

9,905

19,867

Bus Only

M32017

Harvard Square @ Brattle St / Eliot St

Harvard Square: Brattle St

9,698

9,907

19,605

Bus and Rapid Transit

M32029

Porter Square Station

Porter Square Station

9,636

9,239

18,875

Bus, Commuter Rail, and Rapid Transit

M32019

CambridgeSide Galleria - CambridgeSide PL @ Land Blvd

CambridgeSide Galleria

9,085

9,548

18,633

None

M32007

Cambridge St - @ Columbia St / Webster Ave

Cambridge St @ Columbia St

8,958

8,371

17,329

Bus Only

M32021

Harvard University Gund Hall @ Quincy St / Kirkland S

Harvard: Gund Hall

8,457

8,539

16,996

Bus Only

M32014

Harvard University Housing - 115 Putnam Ave @ Peabody Terrace

Harvard Housing: Putnam Ave

8,695

7,870

16,565

None

M32013

Cambridge Main Library @ Broadway / Trowbridge St

Cambridge Main Library

8,301

8,241

16,542

Bus Only

M32016

Harvard Kennedy School @ Bennett St / Eliot St

Harvard Kennedy School

7,210

7,697

14,907

Bus Only

M32026

359 Broadway - Broadway @ Fayette Street

359 Broadway

7,751

6,764

14,515

Bus Only

M32023

Harvard University / SEAS Cruft-Pierce Halls @ 29 Oxford St

Harvard: SEAS Cruft-Pierce Halls

6,939

6,957

13,896

None

M32027

Binney St / Sixth St

Binney St @ Sixth St

6,472

6,428

12,900

None

M32032

Kendall Street

Kendall Street

6,041

6,066

12,107

None

M32038

Harvard University River Houses @ DeWolfe St / Cowperthwaite St

Harvard: River Houses

5,853

5,819

11,672

None

M32030

Dana Park

Dana Park

5,893

5,402

11,295

Bus Only

M32020

Harvard Law School @ Mass Ave / Jarvis St

Harvard Law School

5,683

5,575

11,258

Bus Only

M32024

Harvard University Radcliffe Quadrangle @ Shepard St / Garden St

Harvard: Radcliffe Quadrangle

5,571

5,488

11,059

Bus Only

M32034

EF - North Point Park

EF - North Point Park

5,359

5,271

10,630

None

M32025

Linear Park - Mass. Ave. @ Cameron Ave.

Linear Park: Mass Ave @ Cameron Ave

5,185

5,003

10,188

Bus Only

M32033

Alewife Station @ Russell Field

Alewife Station

3,194

3,322

6,516

Bus Only

M32031

Danehy Park

Danehy Park

2,564

2,517

5,081

None

M32036

Rindge Avenue - O'Neill Library

Rindge Ave

1,289

1,349

2,638

Bus Only

M32041

MIT Pacific St @ Purrington St

MIT: Pacific St

1,029

943

1,972

None

M32042

MIT Vassar St

MIT: Vassar St

766

709

1,475

Bus Only

M32040

University Park

University Park

748

667

1,415

Bus Only

M32039

Lesley University

Lesley Univ

236

264

500

Bus, Commuter Rail, and Rapid Transit

M32043

Mt Auburn

Mt Auburn

207

196

403

Bus Only

Somerville Stations:

 

 

S32006

Davis Square

Davis Square

10,027

11,662

21,689

Bus and Rapid Transit

S32002

Union Square - Somerville

Union Square: Somerville

6,122

6,460

12,582

Bus Only

S32004

Conway Park - Somerville Avenue

Conway Park

5,321

5,242

10,563

Bus Only

S32003

Beacon St @ Washington / Kirkland

Beacon St @ Washington St

4,617

4,375

8,992

Bus Only

S32005

Wilson Square

Wilson Square

3,756

3,521

7,277

Bus Only

S32011

Teele Square @ 239 Holland St

Teele Square

3,258

2,108

5,366

Bus Only

S32010

Somerville Hospital @ Highland Ave / Crocker St

Somerville Hospital

2,387

1,670

4,057

Bus Only

S32001

Somerville City Hall

Somerville City Hall

2,051

1,781

3,832

Bus Only

S32009

Packard Ave / Powderhouse Blvd

Packard Ave @ Powderhouse Blvd

1,861

1,780

3,641

None

S32012

Summer St @ Cutter St

Summer St @ Cutter St

1,563

1,773

3,336

Bus Only

S32008

Powder House Circle - Nathan Tufts Park

Powder House Circle

1,699

1,589

3,288

Bus Only

S32013

Magoun Square @ Trum Field

Magoun Square

1,266

1,349

2,615

Bus Only

Data Source: 2015 Hubway trip logs (4/17/2015-12/18/2015) and CTPS GIS analysis.

a Data for TD Garden - Causeway at Portal Park #1 reflects data for both the TD Garden - Causeway at Portal Park #1 and the TD Garden and the TD Garden - Causeway at Portal Park #2 stations. These stations were combined given their close proximity. b Data for Old State House reflects data for both the Old State House and the Mayor Martin J Walsh - 28 State St. stations. These stations were combined given their close proximity.

CTPS = Central Transportation Planning Staff. O-D = Origin-Destination.

 

Appendix C:
Adjusting Transit Travel Time Estimates to Mitigate Bias

As discussed in Section 4.1, CTPS sought a strategy to mitigate the biases that arise when modeling and calculating the travel time of alternative transit trips so that they can be compared to Hubway trips. The Hubway trip data used in this analysis only included the portion of each Hubway trip from when a rider picked up a bicycle at the origin Hubway station to when she dropped it off at the destination Hubway station. The rider’s true origin (for example, her home), her true destination (for example, her workplace), and the distance she traveled between these locations and Hubway stations are all unknown. The Hubway station coordinates were the only data that CTPS could use to identify the locations associated with a Hubway trip and to model an alternative transit trip. However, using the Hubway stations as trip start-and-end points systematically underestimates the total time of the Hubway trip. Also, using these stations as start-and-end points for an alternative transit trip may overestimate or underestimate travel time by transit, depending on the locations of the rider’s true origin and destination.

 

Figure C-1 illustrates a scenario where setting Hubway trip origins and destinations at Hubway stations overestimates the travel time of the alternative transit trip.

 

FIGURE C-1
Scenario 1: Underestimating Hubway
Travel Time and Overestimating Transit Travel Time

FIGURE C-1: Scenario 1: Underestimating Hubway Travel Time and Overestimating Transit Travel Time: This figure depicts how using Hubway stations as proxies for true trip origins and destinations (without any other adjustments) may underestimate total travel time to complete the trip via walking and Hubway, while overestimating the time it would take to complete the trip via walking and transit.

 

In scenario shown in Figure C-1, the true origin and true destination locations are closer to transit stations than to Hubway stations. In the diagram, the solid green arrow corresponds to the recorded Hubway trip between the two stations, while the dashed green arrows represent the trips of unknown distance that the rider would need to make from her origin to the first Hubway station and from the second Hubway station to her true destination. The blue arrows represent an estimated alternative transit trip that starts and ends at the Hubway stations, and includes walk trips between the Hubway stations and transit stations. The dashed purple arrows represent a transit alternative for making the complete trip that does not involve stopping at Hubway stations. In this hypothetical situation, by using Hubway stations as proxies for the true origin and destination, CTPS underestimated the total Hubway trip time—because the trips to and from Hubway stations are not included—and overestimated the amount of time needed to access and egress transit, thereby overestimating total transit travel time.

 

Figure C-2 illustrates a scenario where setting Hubway trip origins and destinations at Hubway stations underestimates the travel time of the alternative transit trip.

 

FIGURE C-2
Scenario 2: Underestimating Hubway
 Travel Time and Underestimating Transit Travel Time

FIGURE C-2: Scenario 2: Underestimating Hubway Travel Time and Underestimating Transit Travel Time: This figure depicts how using Hubway stations as proxies for true trip origins and destinations (without any other adjustments) may underestimate total travel time to complete the trip via walking and Hubway, while also underestimating the time it would take to complete the trip via walking and transit.

 

In this hypothetical case, the rider’s true origin and true destination are closer to Hubway stations than to transit stations. Using Hubway stations to represent the origin and destination underestimates the amount of time needed to access and egress transit stations, and thereby underestimates total transit travel time. Also, as in the first scenario, the Hubway trip does not account for the walk access and walk egress trips between the true origin and destination and Hubway stations, which would lead to underestimating total Hubway travel time. 

 

For this analysis described in Section 4, CTPS assumed that the actual start and end points of Hubway trips have a random geographic distribution with respect to Hubway stations and transit stations. If the trips represented by the Hubway data are considered in aggregate, the sums of the variations in walking time to and from Hubway stations (dashed green arrows) would be cancelled out by the sums of the variations in walking time to and from transit stations (dashed purple arrows). Using this assumption, CTPS removed walk access and walk egress time from OTP’s transit travel time estimates, in order to compare a Hubway trip’s travel time with the travel time of OTP’s transit alternative more fairly. Figure C-3 illustrates the effect of CTPS’s approach.

 

FIGURE C-3
Results of CTPS Approach

FIGURE C-3: Results of CTPS Approach: This figure depicts the approach the Central Transportation Planning Staff (CTPS) used to address the bias that may be introduced by using Hubway stations as proxies for true trip origins and destinations. CTPS assumed trips origins and destinations follow a random geographic distribution with respect to Hubway and transit stations. CTPS also removed the time to walk between Hubway stations and transit stations from estimates of transit travel time.

 

This approach has a noticeable quantitative impact on estimates of transit travel time, as walk access and egress time make up about 38 percent of all transit travel time for OTP’s recommended transit alternatives to Hubway trips, on average. However, in the absence of more detailed trip information, CTPS believes this is the best option to correct for the systematic bias introduced by using Hubway stations as proxies for actual trip origins and destinations.

 

Appendix D:
Methods used to Estimate
Uncomfortable Conditions

CTPS used the following resources to identify evidence of uncomfortable conditions on alternative transit routes identified in Section 4.6, “Results for Origin Destination Pairs with Large Shares of Trips that are Faster or Comparable by Transit”:

 

Point-Checks for MBTA Bus Routes

Data tabulators count bus passengers at the maximum load point for bus trips traveling either inbound or outbound during peak periods. These data-collection activities are referred to as point checks. If the passenger load at the point check falls between 100 and 140 percent of seated bus capacity, the trip is considered overburdened. If the load at the point check is greater than 140 percent of seated capacity, the trip is considered overcrowded. CTPS reviewed fall 2015 point-check data for bus routes that could provide substitute service for a Hubway O-D pair. If the point-check location fell within the bus route segment that would serve as a transit alternative for a particular Hubway O-D pair, CTPS counted the number and share of overcrowded or overburdened bus trips during the relevant peak period. If the point-check data showed that more than 50 percent of trips for a given route were overcrowded or overburdened during the peak period (AM or PM), then CTPS considered this information to be evidence of uncomfortable conditions on that route.

 

Percentage of Passenger Time in Comfortable Conditions on MBTA Bus Routes 

To understand passenger comfort levels on MBTA bus routes better, the MBTA has calculated the ratio of comfortable passenger time to total passenger time for each route during the course of the day using data for average weekdays in fall 2015 from its busload dataset.46 This data—which is in active development and is considered experimental—combines average trip loads from on-bus automated passenger counter (APC) data and information on boarding and alighting locations from the MBTA’s ODX model. During high-volume periods, whenever passenger loads exceeded 140 percent of seated capacity, then all passengers were considered as being uncomfortable.47

 

In its current Service Delivery Policy, the MBTA set a minimum threshold of 92 percent of passenger minutes in comfortable conditions for its bus routes.48 CTPS queried bus crowding data from fall 2015 for the MBTA’s relevant bus routes and directions (inbound or outbound) during the peak periods (when information was available), and documented the percentage for the overall routes, as this ratio is the most meaningful at the route level. When a relevant bus route failed to meet the minimum threshold (92 percent) during the relevant peak period, CTPS noted this as possible evidence of uncomfortable conditions on that bus route. CTPS also gathered information for relevant route segments that that would support transit alternatives for Hubway O-D pairs during the AM and PM peak periods. When a bus route failed to meet the minimum threshold during these periods on at least one part of these segments, then CTPS noted this as possible evidence of uncomfortable conditions on that segment.

 

Rapid Transit System Flow Data

CTPS reviewed MBTA disaggregate automated fare collection data from May 10–12, 2016.49 This data had been processed through CTPS’s CCARD2OD model, which infers origins and destinations on the rapid transit system. This data showed the average passenger load on trains departing from rapid transit stations, by line, for each hour of the day.

 

To determine whether loads at specific rapid transit stations might indicate uncomfortable conditions, CTPS estimated the passenger capacity per hour for specific rapid transit lines. The passenger capacity per train was estimated using standards for total passenger vehicle loads on rapid transit during peak periods, which are listed in the MBTA’s Service Delivery Policy. These vehicle loads were multiplied by an estimated number of cars per train, as shown in Table D-1. 

 

 


 

TABLE D-1
Estimated Passenger Capacity for
Rapid Transit Trains during Peak Periods

Line

Total Passenger
Load Capacity

per Vehicle  

Estimated Cars per Train

Estimated Total Passenger Load Capacity per Train

Red

165

6

990

Orange

141

6

846

Blue

86

6

516

Green

100

2

200

Data source for passengers per vehicle: MBTA, MBTA Service Delivery Policy, p. 45.   

 

CTPS then multiplied these values by an estimated number of trains per hour on different lines (accounting for direction) of the rapid transit system, as shown in Table D-2.

 

TABLE D-2
Estimated Passenger Capacity per Hour on Rapid Transit

 

 

 

 

 

Line

Segment

Trains Per Hour (All Relevant Lines)

Estimated Total Passenger  Load Capacity per Train

Estimated Passenger Load Capacity Per Hour

Red

Alewife to JFK U-Mass (Ashmont, Braintree)

13

990

13,200

Red

JFK U-Mass to Braintree

7

990

6,600

Red

JFK U-Mass to Ashmont

7

990

6,600

Red

Braintree to JFK/U-Mass

7

990

6,600

Red

Ashmont to JFK/U-Mass

7

990

6,600

Red

JFK/UMass to Alewife (Ashmont, Braintree)

13

990

13,200

Orange

Oak Grove to Forest Hills

10

846

8,460

Orange

Forest Hills to Oak Grove

10

846

8,460

Green

Lechmere to North Station (E )

10

200

2,000

Green

North Station to Park St (C, E)

20

200

4,000

Green

Park Station to Copley (B, C, D, E)

40

200

8,000

Green

Copley to Kenmore (B, C, D)

30

200

6,000

Green

Copley to Symphony (E)

10

200

2,000

Green

Surface B Line to Boston College

10

200

2,000

Green

Surface C Line to Cleveland Circle

10

200

2,000

Green

Surface D Line to Riverside

10

200

2,000

Green

Surface E Line to Heath Street

10

200

2,000

Green

North Station to Lechmere (E)

10

200

2,000

Green

Park to North Station ( C, E)

20

200

4,000

Green

Copley to Park (B, C, D, E)

40

200

8,000

Green

Kenmore to Copley (B, C, D)

30

200

6,000

Green

Symphony to Copley (E)

10

200

2,000

Green

Surface B Line to Kenmore

10

200

2,000

Green

Surface C Line to Kenmore

10

200

2,000

Green

Surface D Line to Kenmore

10

200

2,000

Green

Surface E Line to Copley

10

200

2,000

Data source for rapid transit service frequencies by line: MBTA Rapid Transit schedule: December 31, 2017 to March 24, 2017, http://www.mbta.com/uploadedFiles/Documents/Schedules_and_Maps/Subway/frequency-schedule.pdf, accessed January 26, 2017, p. 2.

Note: These frequencies were used to calculate the trains per hour, which have been rounded to the nearest whole number of trains in this table.

 

CTPS noted that there could be uncomfortable conditions on trains departing from a particular station at a particular hour if the ratio between the average load at that station and the total passenger capacity of the relevant rapid transit line was greater than 0.6 for that hour. This threshold was selected because at this level, the passenger loads would be at least 125 percent of seated capacity on all transit lines. It also allows for some variations in passenger loads on individual trains that may pass through a station during a one-hour period. CTPS noted that there could be uncomfortable conditions on a rapid transit segment that would provide an alternative for a Hubway O-D pair if that segment included at least one station meeting the 0.6 ratio threshold during at least one hour of the AM or PM peak period.

 

Commuter Rail Data

In 2012, CTPS conducted passenger counts for the MBTA commuter rail system, which were used to determine line volumes. The commuter rail segment between South Station and Back Bay Station—served by the Framingham/Worcester, Franklin, Needham, and Providence/Stoughton Lines—is the only option that emerged as an alternative for the Hubway O-D pairs analyzed in Section 4.6. Based on 2012 passenger count data, on these rail lines, the maximum loads occur at some point prior to Back Bay inbound and after Back Bay outbound. By comparison, the commuter rail traffic between Back Bay and South Station is low. For the purposes of this analysis, CTPS assumed that crowding did not take place on this segment, although crowding could theoretically occur because of non-uniform distribution of passengers between cars on a train.

 

Appendix E:
Details on O-D Pairs with Large Shares of Trips Faster
or Comparable by Transit

 

 

 


TABLE E-1
O-D Pairs With Large Shares of Trips Faster or Comparable by Transit (AM Peak Period)

 

 

 

 

 

 

 

 

 

Evidence of
Uncomfortable Conditions
a

O-D Pair: Station Numbers

O-D Pair:
Station Names

Total Weekday AM Peak Period Trips

Percent Trips Faster or Comparable by Transit

Main Alternative Transit Mode for Trips Faster or Comparable by Transit

Transit Mode Share of All O-D Pair Trips

 Transit Mode Share of Trips Faster or Comparable by Transit

Hubway: Potential Complement or Potential Competitor?

Transit Lines and
Segments

Bus Point Checks

Bus Route Comfort Fraction Data

Bus Segment Comfort Fraction Data

Loads at Rapid Transit Stations

D32022 to C32007

TD Garden  to Prudential Center

109

98%

Rapid Transit (0 Transfers)

100%

100%

Complement

1. Green Line (North Station to Prudential)

 

 

 

X

 

 

 

 

 

2. Orange Line (No Stan to Massachusetts Avenue)

 

 

 

X

D32022 to C32003

TD Garden to Back Bay / South End

128

86%

Rapid Transit (0 Transfers)

100%

100%

Complement

1. Orange Line (No Sta to Back Bay)

X

2. Green Line (No Sta to Copley)

X

D32022 to D32007

TD Garden  to Boylston St @ Arlington St

139

83%

Rapid Transit 0 Transfers)

100%

100%

Complement

1. Orange Line (No Sta to Back Bay)

 

 

 

X

 

 

 

 

 

2. Green Line (No Sta to Arlington)

 

 

 

X

D32005 to A32010

Boston Public Library to South Station

146

63%

Commuter Rail (0 Transfers)

100%

100%

Competitor

1. Multiple Lines (Back Bay to So Sta)

D32022 to C32008

TD Garden to Boylston St @ Fairfield St

178

61%

Rapid Transit (0 Transfers)

100%

100%

Complement

2. Green Line (No Sta to Copley)

 

 

 

X

M32022 to A32004

Lower Cambridgeport to Longwood Ave @ Binney St

111

58%

Bus (0 Transfers)

100%

100%

Complement

1. Route 47 (Magazine St @ Riverside Rd to Brookline Ave @ Longwood Ave)

X

X

A32010 to D32018

So Sta to BCEC

305

54%

Bus (0 Transfers)

100%

100%

Competitor

1.Route 7 bus (So Sta to Summer St Op WTC Avenue)b

 

 

 

 

 

 

 

 

 

2.Route 4 (So Sta to Summer St Op WTC Ave)c

 

 

 

 

Data sources: 2015 Hubway trip logs (4/17/201512/18/2015), Open Trip Planner and Google Maps output, MBTA fall 2015 bus point-check data, MBTA fall 2015 bus crowding data, MBTA AFC data (May 10-12, 2016),

a An "X" in the Evidence of Uncomfortable Conditions columns indicates that CTPS identified evidence in this category, according to the methods described in Appendix D. b Point-check data was not available for Route 7 buses traveling outbound during the AM Peak period. c Point-check data and bus route comfort fraction data were not available for Route 4 buses traveling outbound during the AM Peak period.

AFC = automated fare collection. BCEC = Boston Convention and Exhibition Center. CTPS = Central Transportation Planning Staff. O-D = Origin-Destination. OTP = Open Trip Planner.

 

TABLE E-2
O-D Pairs With Large Shares of Trips Faster or Comparable by Transit (PM Period)

 

 

 

 

 

 

 

 

 

Evidence of
Uncomfortable Conditions a

O-D Pair: Station Numbers

O-D Pair:
Station Names

Total Weekday PM Peak Period Trips

Percent of Trips Faster or Comparable by Transit

Main Alternative Transit Mode for Trips Faster or Comparable by Transit

Transit Mode Share of All O-D Pair Trips

Transit Mode Share of Trips Faster or Comparable by Transit

Hubway: Potential Complement or Potential Competitor?

Transit Lines and Segments

Bus
Point Checks

Bus Route Comfort Fraction Data

Bus Segment Comfort Fraction Data

Loads at Rapid Transit Stations

C32007 to D32022

Prudential Center to TD Garden

194

98%

Rapid Transit (0 transfers)

100%

100%

Complement

1. Green Line (Prudential to North Station)

X

2. Orange Line  (Massachusetts Avenue to No Sta)

X

A32010 to C32003

South Station to Back Bay/South End

128

96%

Commuter Rail Only (0 transfers)

98%

100%

Competitor

1. Commuter Rail (Multiple Lines, So Sta to Back Bay)

 

 

 

 

A32010 to C32012

So Sta to Andrew Sta

104

91%

Rapid Transit (0 transfers)

100%

100%

Complement

1. Red Line (So Sta to Andrew)

X

C32003 to A32010

Back Bay /So End to So Sta

204

91%

Commuter Rail (0 transfers)

95%

100%

Competitor

1. Commuter Rail (Multiple Lines, Back Bay to So Sta)

 

 

 

 

D32014 to D32005

Tremont St @ West St to Boston Public Library

110

81%

Rapid Transit (0 transfers)

98%

100%

Complement

1. Green Line (Park St to Copley)

X

2. Orange Line (Chinatown to Back Bay Sta)

X

D32009 to D32021

Faneuil Hall to Charlestown: Warren St

102

78%

Bus (0 transfers)

100%

100%

Complement

1. Route 93 (Congress St @ North St to Chelsea St @ Warren St)

 

X

X

 

 

 

 

 

 

 

2. Route 92 (Congress St @ North St to Warren St @ Winthrop St)b

 

 

 

 

C32003 to D32022

Back Bay/ So End to TD Garden

141

75%

Rapid Transit (0 transfers)

100%

100%

Complement

1. Orange Line (Back Bay to No Sta)

X

A32010 to D32005

So Sta to Boston Public Library

167

61%

Commuter Rail Only (0 transfers)

98%

100%

Competitor

1. Commuter Rail (Multiple Lines, So Sta to Back Bay)

 

 

 

 

D32007 to D32022

Boylston St @ Arlington St to TD Garden

333

58%

Rapid Transit (0 transfers)

100%

100%

Complement

1. Green Line (Arlington to No Sta)

X

A32003 to A32012

BU Central to Packard's Corner

168

55%

Bus (0 transfers)

90%

100%

Complement

1. Route 57 (Commonwealth Ave @ University Rd to 1079 Comm Ave)

X

X

X

 

 

 

 

 

 

 

2. Route 57A (Comm Ave @ University Rd to 1079 Comm Ave)

X

X

X

 

A32010 to D32018

So Sta to BCEC

156

55%

Bus (0 transfers)

100%

100%

Complement

1. Route 7 bus (So Sta to Summer St op WTC Ave)

X

X

2. Route 4 bus (So Sta to Summer St op World Trade Center Ave)c

M32011 to M32018

Central Square to Harvard Sq: Mass Ave @ Dunster St

119

54%

Rapid Transit (0 transfers)

86%

94%

Complement

1. Red Line (Central to Harvard)

 

 

 

X

B32007 to A32010

Seaport Sq to So Sta

105

53%

Bus (0 transfers)

100%

100%

Complement

1.SL1 (Courthouse to So Sta)d

X

2. SL2 (Courthouse to So Sta)

X

X

 

 

 

 

 

 

3. SL Waterfront (Courthouse to So Sta)e

 

 

 

 

B32010 to A32002

Kenmore Sq to Agganis Arena

124

53%

Bus (0 transfers)

92%

98%

Complement

1. 57 (Kenmore to Comm Ave @ Buick St)

X

X

X

 

 

 

 

 

 

 

2. 57/57A (Kenmore to Comm Ave @ Buick St)

X

X

X

 

D32012 to D32021

Post Office Sq to Charlestown: Warren St

138

51%

Bus (0 transfers)

100%

100%

Complement

1. Route 93 (Pearl St @ Congress St to Chelsea St @ Warren St)

X

X

2. Route 92 (Pearl St @ Congress St to Warren St @ Winthrop St)b

M32001 to D32022

Lechmere Sta to TD Garden

255

51%

Rapid Transit (0 transfers)

59%

100%

Competitor

1. Green Line (Lechmere to No Sta)

 

 

 

 

Data sources: 2015 Hubway trip logs (4/17/201512/18/2015), Open Trip Planner and Google Maps output, MBTA fall 2015 bus point-check data, MBTA fall 2015 bus crowding data, MBTA AFC data (May 10-12, 2016),

a An "X" in the Evidence of Uncomfortable Conditions columns indicates that CTPS identified evidence in this category, according to the methods described in Appendix D. b Comfort fraction data was not available for Route 92 buses traveling outbound during the AM Peak period. c Point-check data and bus route comfort fraction data were not available for Route 4 buses traveling outbound during the PM Peak period. d Comfort fraction data was not available for SL 1 buses traveling inbound during the PM peak period. eComfort fraction data was not available for SL Waterfront buses traveling inbound during the PM peak period. 

AFC = automated fare collection. BCEC = Boston Convention and Exhibition Center. BU = Boston University. CTPS = Central Transportation Planning Staff. O-D = Origin-Destination. OTP = Open Trip Planner.

 

 

 

TABLE E-3
O-D Pairs With Large Shares of Trips Faster or Comparable by Transit (Weekday Off-Peak Periods)

 

 

 

 

 

 

Alternative Transit Modes and Transfers
for Trips Faster or Comparable by Transit

 

 

 

 

 

 

Bus

Rapid
Transit

Commuter Rail

O-D Pair: Station Numbers

O-D Pair:
Station Names

Weekday
Time Period

Total Trips

Number of
Trips Faster or Comparable by Transit

Percent of Trips Faster or Comparable by Transit

Zero
Transfers

One
Transfer

Zero
Transfers

One
Transfer

Zero
Transfers

D32022 to D32007

TD Garden  to Boylston St @ Arlington St

Late Night / Early Morning

113

106

94%

0%

0%

100%

0%

0%

A32003 to A32012

BU Central to Packard's Corner

Midday

113

72

64%

97%

0%

3%

0%

0%

M32018 to M32011

Harvard Square: Mass Ave @ Dunster St to Central Sq

Midday

110

86

78%

1%

0%

99%

0%

0%

M32011 to M32018

Central Square to Harvard Sq: Mass Ave @ Dunster St

Midday

109

61

56%

10%

0%

90%

0%

0%

K32003 to A32008

Brookline Village to Buswell St @ Park Dr

Midday

100

89

89%

0%

0%

100%

0%

0%

A32003 to A32012

BU Central to Packard's Corner

Evening / Night

125

72

58%

97%

0%

3%

0%

0%

M32011 to M32018

Central Square to Harvard Sq: Mass Ave @ Dunster St

Evening / Night

121

63

52%

32%

0%

68%

0%

0%

M32016 to A32019

Harvard Kennedy School to Harvard: Transportation Services

Evening / Night

101

77

76%

100%

0%

0%

0%

0%

Data sources: 2015 Hubway trip logs (4/17/201512/18/2015) and OTP output.

BU = Boston University. O-D = Origin-Destination. OTP = Open Trip Planner.

 

TABLE E-4
O-D Pairs With Large Shares of Trips Faster or Comparable by Transit (Weekend)

 

 

Alternative Transit Modes and Transfers
for Trips Faster or Comparable by Transit

Bus

Rapid Transit

Commuter Rail

O-D Pair: Station Numbers

O-D Pair:
Station Names

Total Trips

Number of
Trips Faster or Comparable by Transit

Percent of Trips Faster or Comparable by Transit

Zero
Transfers

One
Transfer

Zero
Transfers

One
Transfer

Zero
Transfers

D32014 to D32005

Tremont St @ West St to Boston Public Library

105

93

89%

0%

0%

100%

0%

0%

M32018 to M32011

Harvard Square: Mass Ave @ Dunster St to Central Square

151

118

78%

3%

0%

97%

0%

0%

A32010 to C32003

South Station to Back Bay / South End

110

72

65%

0%

0%

7%

0%

93%

A32010 to C32017

South Station to South Boston Library

117

75

64%

96%

4%

0%

0%

0%

M32011 to M32018

Central Square to Harvard Square: Mass Ave @ Dunster St

176

108

61%

16%

0%

84%

0%

0%

A32003 to A32012

BU Central to Packard's Corner

151

87

58%

95%

0%

5%

0%

0%

M32018 to M32029

Harvard Square: Mass Ave @ Dunster St to Porter Square Station

122

70

57%

43%

0%

57%

0%

0%

D32022 to M32001

TD Garden  to Lechmere Station

102

58

57%

0%

0%

100%

0%

0%

M32016 to A32019

Harvard Kennedy School to Harvard: Transportation Services

107

56

52%

100%

0%

0%

0%

0%

B32010 to A32002

Kenmore Square to Agganis Arena

119

61

51%

98%

0%

2%

0%

0%

A32010 to C32006

South Station to Washington St @ Rutland St

119

60

50%

100%

0%

0%

0%

0%

B32008 to D32022

Old State House to TD Garden

152

76

50%

0%

0%

100%

0%

0%

A32010 to D32018

South Station to BCEC

151

75

50%

100%

0%

0%

0%

0%

M32010 to M32018

Inman Square to Harvard Square: Mass Ave @ Dunster St

105

52

50%

100%

0%

0%

0%

0%

Data sources: 2015 Hubway trip logs (4/17/201512/18/2015) and OTP output.

BCEC = Boston Convention and Exhibition Center. BU = Boston University. O-D = Origin-Destination. OTP = Open Trip Planner.

 


Appendix F:
Details on O-D Pairs with Large Shares of Trips
Much Slower by Transit
50

 

 


TABLE F-1
O-D Pairs with 75 Percent or More Trips Much Slower by Transit
(Weekday AM Peak Period)

 

 

Alternative Transit Modes and Transfers for Trips Much Slower by Transit

 

 

Bus

Rapid Transit

Bus and
 Rapid Transit

Commuter
Rail

Other

O-D Pair: Station Numbers

O-D Pair:
Station Names

Total Trips

Number of Trips Much Slower by Transit

Percent of Trips Much Slower by Transit

Zero
Transfers

One
Transfer

Zero
Transfers

One
Transfer

One
Transfer

Two
Transfers

Zero
Transfers

One
Transfer

Two
Transfers

D32006 to C32010

Lewis Wharf to Congress St @ Sleeper St

135

135

100%

84%

1%

0%

2%

13%

0%

0%

0%

0%

C32024 to D32012

State Street to Post Office Square

106

105

99%

0%

0%

100%

0%

0%

0%

0%

0%

0%

B32004 to C32024

Aquarium Station to State Street

161

158

98%

0%

4%

0%

1%

95%

0%

0%

0%

0%

K32001 to B32003

Coolidge Corner to Harvard: HMS/HSPH

216

209

97%

99%

0%

0%

0%

1%

0%

0%

0%

0%

A32010 to D32017 a

South Station to The Esplanade

101

95

94%

6%

0%

0%

24%

0%

0%

0%

69%

0%

K32004 to A32004

JFK Crossing to Longwood Ave @ Binney St

102

95

93%

63%

37%

0%

0%

0%

0%

0%

0%

0%

A32017 to B32003

Allston Green District to Harvard: HMS/HSPH

179

166

93%

78%

0%

0%

0%