As of September 1, 2012, the University Transportation Center for Mobility (UTCM) is no longer an active center of the Texas A&M Transportation Institute. The archived UTCM website remains available here.

UTCM HomeConference Site MapContact Conference Webmaster

Conference on Performance Measures for Transportation and Livable Communities

SEPTEMBER 7-8, 2011 • AUSTIN, TEXAS

BREAKOUT SESSION 7:
Access to Destinations Performance Measures

Click on the icon to view a PDF of the slide presentation.

Presiding

GINA BAAS, Center for Transportation Studies, University of Minnesota

	An Open-Source Approach to Calculating Accessibility MIKE McGURRIN, Noblis, Inc.
	Access to Destinations: Methods, Findings, and Implications MICHAEL IACONO, University of Minnesota
	Impact of Light Rail Implementation on Labor Markets Accessibility: A Transportation Equity Perspective ANDREW GUTHRIE, Humphrey School of Public Affairs, University of Minnesota
	Accessibility Ratings ANNA GARTSMAN and JESSICA CASEY, Dukakis Center for Urban and Regional Policy, Northeastern University

---

An Open-Source Approach to Calculating Accessibility

MIKE McGURRIN, Noblis, Inc.

Slides [PDF , 6.2M]

Mike McGurrin discussed the use of open government data and open-source software to develop simple low-cost analysis tools and visualization techniques.  He described examples using available data, challenges, and future opportunities.  Mike covered the following topics in his presentation.

• The movement of people and goods by different elements of the transportation system is a means to an end, not an end in and of itself.  The transportation system supplies goods or services, providing access to jobs, and allows families to reach vacation destinations.  Accessibility metrics measure the ease of reaching the desired destinations.  They measure transportation outcomes, rather than intermediate performance or problems.  Accessibility metrics have been recommended by multiple organizations.  Accessibility metrics should be rigorous, quantifiable, transparent, and understandable.  They have previously been difficult to analyze.
• A new approach to analyzing accessibility is to utilize the growing set of open government data and open-source software to develop simple low-cost analysis tools and visualizations.  A data appender and/or visualization tools can be used as stand-alone techniques to visualize results.  Input is simple using XML-based files.
• One example of using this approach is the on-time performance (OTP) travel times, which are calculated using data from the Washington Metropolitan Area Transit Authority (WMATA).  Comparison of OTP travel times with the WMATA trip planner indicates similar results.  Visualization tools and techniques can also be used to display data available from public agencies and other open sources.
• One challenge with the use of existing tools and data is the long run times needed on a single processor.  For example, it can take two days to run 250,000 trip calculations in 500 regions.  Potential solutions to this challenge include separating out the walk time calculations to avoid a huge duplicative effort.  It also greatly simplifies the remaining graph and travel time analysis.  Parallel processing represents another possible solution, as the problem is ideally suited for splitting over multiple processors.
• Another challenge is that many agencies do not publish transit schedules in General Transit Feed Specification (GTFS) format.  For example, WMATA, Arlington County, and the DC Connector do publish transit schedules in a GTFS format, but Fairfax and Loudon Counties do not.  Some GTFS formats may be available upon special request, however.  It is labor intensive, but it is possible to create schedules in the GTFS format from other formats.
• An additional challenge is that there is no open data on automobile travel times.  The development of connected vehicles offers a potential opportunity for anonymous, crowd-sourced open travel time data.  If this data becomes available it would open up traveler information applications in the same way that the GTFS opened up transit information applications.
• In conclusion, new approaches using open government data and open-source software can reduce the investment costs and level of effort required to conduct such analyses.  There is still a dependence on having the necessary data, however.  Further work is needed to reduce run-times for large networks.  Tools are available to provide easily used and understood visualizations to communicate results.  The source code is available as an open-source code at Open Source Accessibility Toolkit (OSAT): https://github.com/Noblis/OSAT.  The source code, example data, sample outputs, and documentation are available.

[ Top ]

Access to Destinations: Methods, Findings, and Implications

MICHAEL IACONO, University of Minnesota

Slides [PDF , 3.0M]

Michael Iacono discussed a major study examining access to destinations in the Minneapolis-St. Paul metropolitan area.  He described the study objectives, the process, and findings.  Michael covered the following topics in this presentation.

• Accessibility is a measure that relates the transportation network to the pattern of activities that comprise land use.  It measures the ease of reaching valued destinations.  It has been suggested that accessibility may be the most important concept in defining and explaining regional form and function.
• The objectives of the access to destinations project included improving the understanding of travel on the current transportation infrastructure, developing measures of accessibility using travel and land use data, and examining how accessibility has changed from 1995 to 2005.  An additional objective was using these new tools and information to assess how the existing transportation and land use system meets alternative policy goals.
• The access to destinations project was sponsored by the Minnesota Department of Transportation (MnDOT), Hennepin County, the Metropolitan Council, and the McKnight Foundation.  Numerous faculty and researchers at the University of Minnesota and the University of Colorado were involved in the project.
• Accessibility can be generalized as more modes, more destinations, and more opportunities.  The ability to reach jobs, schools, parks, shopping, and other destinations by automobile, transit, bicycling, and walking can also be measured or modeled.
• Public meetings were held to help obtain input on possible accessibility performance measures.  Existing literature suggests that accessibility performance measures should be clear, comprehensive, cumulative, comparable, and calculable.  For example, accessibility can be thought of as cumulative opportunities.  Accessibility can be defined by the number of jobs, workers, or other destinations, that can be reached in different lengths of time by different modes during different times of the day.
• A number of different data sources were used in the analysis.  These sources focused on travel time information, the number of workers, the number of jobs, and origin and destination data.  Data from the U.S. Census, the Metropolitan Council, the Center for Urban and Regional Affairs at the University of Minnesota, the State of Minnesota, and MnDOT were used in the study.  The analysis highlighted that modeling accessibility is relatively straight-forward, but measuring accessibility is more difficult.
• Travel time data for the Minneapolis-St. Paul metropolitan area freeway network was generated using the system of embedded speed and volume sensors.  The resulting speeds were displayed graphically on maps of the metropolitan area.
• The arterial travel time estimation process used in the study included a number of steps.  The first step was to assign an origin/destination matrix using SUE assignment.  The second step was to correct flows according to observed Automatic Traffic Recording (ATR) station counts, where available.  The third step was to apply corrections based on covariance of links without ATR stations to links with ATR stations.  The fourth step was to use the link performance function to estimate travel time based on corrected flows.  Obtaining travel speed data on freeways and arterials will be easier in the future due to vehicles equipped with global positioning satellite (GPS) devices.  An initial analysis of mean speed estimates from GPS data was conducted as part of the study.
• The study also examined if accessibility in the metropolitan area was increasing or decreasing over time.  This analysis examined the change from 1995 to 2005 in the number of jobs accessible from origin traffic analysis zones by automobiles in 10 minutes, 20 minutes, and 30 minutes during the morning peak hour.  The analysis indicates that accessibility may be decreasing for short trips due to congestion, but increasing for long trips where land use outweighs congestion.
• Access to jobs by other modes, including transit, walking, and bicycling was examined in the study.  The number of jobs accessible within 20 minutes by these modes were estimated and presented graphically.
• The difference between accessibility and congestion was also examined.  The delay per passenger during the morning peak period for the metropolitan area from the TTI Urban Mobility Report increased from 19 hours in1990 to 43 hours in 2000.  Accessibility to residents from downtown Minneapolis increased by approximately 15 percent between 1990 and 2000, however.
• Possible policy implications were examined as part of the study.  For example, alternative goals focusing on accessibility were identified.  These goals included maximizing overall regional accessibility, maximizing accessibility for those who are least advantaged, and ensuring choices. 
• Access is a publicly or socially provided benefit to private landowners.  The study examined the potential to capture this value to finance infrastructure which creates access.  This analysis is documented in the Value Capture for Transportation Finance:  Technical Research Report.
• Additional information on the Access to Destinations research project is available at http://cts.umn.edu/access-study/ and http://nexus.umn.edu.

[ Top ]

Impact of Light Rail Implementation on Labor Markets Accessibility: A Transportation Equity Perspective

ANDREW GUTHRIE, Humphrey School of Public Affairs, University of Minnesota

Slides [PDF , 1.2M]

Andrew Guthrie discussed a study examining the impact of the new Hiawatha LRT line in the Minneapolis-St. Paul metropolitan area on labor market accessibility.  He described the Hiawatha LRT line, the study findings, and possible policy implications.  Andrew covered the following topics in his presentation.

• The study examined the locations of low-wage workers and jobs relative to LRT and bus connections to assess if a spatial mismatch exists.  The study included a before-and-after comparison using 2002-2006 Longitudinal Employer-Household Dynamics (LEHD) data.  The general residence locations of low-wage workers and employment sites of low-wage workers were identified and presented geographically.
• Accessibility was defined as the total number of low-wage jobs reachable within 30 minutes of travel by transit.  The study area was the transit service area.  The LEHD 2002-2006 database was used in the analysis.  Overall, the accessibility for low-income jobs within 30 minutes improved after the Hiawatha LRT line opened.
• The Hiawatha LRT line runs from the Mall of America to downtown Minneapolis.  The line serves the Minneapolis-St. Paul International Airport, the Veterans Administration Hospital complex, the Metrodome, and other destinations in the corridor.  The line is 12 miles in length and includes 17 stations.  Connecting bus routes are provided at many stations.  The Hiawatha line, which opened in 2004, is the first LRT line in the Minneapolis-St. Paul metropolitan area.
• The analysis also examined transit area types, including station areas, connection areas, and regular transit-served areas.  Connection areas provide direct bus-LRT connections, with LRT reachable within 30 minutes.  Approximately 22 percent of all the metropolitan area population lives within connected areas.
• Changes in low-wage job accessibility across the different areas were examined by different time periods.  Accessibility improved the most in the station areas for all time periods.  Improvements in accessibility were also realized in the connected areas, and to a lesser extent in the regular service area.  The accessibility changes during the morning peak hours from 7:00 a.m. to 8:00 a.m. were examined in more detail.  The results were displayed graphically, illustrating the improvement in station areas and connecting areas.
• An analysis of before-and-after commuter flows was also conducted.  This analysis estimated changes in the home-to-work flow volume.  This analysis indicated that increased accessibility does not necessarily result in actual gains in low-income commuter flows.  The job accessibility of a neighborhood or the number of job opportunities reachable by workers living in the neighborhood influences the actual results.  For increased accessibility to equal actual gains, workers living in the station and connection areas commute to jobs in the station and connection areas.  The workers moving in and the jobs moving in to the downtown areas, north Hiawatha Station areas, south Hiawatha Station areas, and suburban areas were examined.  The downtown station areas had the highest number of jobs moving in and the north Hiawatha Station areas had the highest number of workers moving in.
• The research identified a number of potential policy implications from the analysis, including those providing positive evidence of LRT’s role in promoting social equity.  One policy implication is that the Hiawatha lines impact is regional in scale.  A second policy implication is that transit system integration and coordination is essential to improving accessibility.  A third policy implication is that low-wage workers and employers are able to take advantage of LRT, at least in the context of the Hiawatha line.  A fourth policy implication associated with land use planning is that balancing jobs and housing can be accommodated at a corridor level, as well as at the site and station levels.

[ Top ]

Accessibility Ratings

ANNA GARTSMAN and JESSICA CASEY, Dukakis Center for Urban and Regional Policy, Northeastern University

Slides [PDF , 939K]

Anna Gartsman and Jessica Casey discussed a study developing and applying accessibility ratings to municipalities in Massachusetts.  They described the study methodology, the objectives, the initial analysis, and future activities.  Anna and Jessica covered the following topics in their presentation.

• Destinations were examined in the study instead of origins for a number of reasons.  First, examining destinations provides a market analysis approach to amenities, but adds a transportation focus.  Second, it allows for the measurement of the number of people who have access to goods and services in each town, instead of how many goods and services that people can access.  The approach can also translate into municipalities altering land use adoption and regulatory control. For example, goods and services can easily be relocated or located to increase coverage and market share, while residential land use can be increased around existing town centers and other highly active regions to increase coverage and accessibility.
• The research focused on a number of objectives.  These objectives included conducting Massachusetts-based research, without LEHD data, opening up the spectrum of destinations beyond jobs, and developing a performance metric for multiple modes of transportation over a diverse geography.  Other objectives included developing a performance metric for aggregating accessibility scores to rank cities and towns in Massachusetts and approaching transportation questions with a severe limitation in data accessibility, without using too many assumptions.
• Some of the limitations encountered with the research included the lack of available updated data and the lack of LEHD data for some areas.  Creative solutions and improvised measures were used to address these limitations.
• Aspects of accessibility include individuals and groups, modes, and activities.  These aspects can be thought of as three overlapping circles.  The analysis focuses on the State of Massachusetts.  The 2000 population of the state was approximately 6.3 million.  There are 205 colleges and universities in the state, and 82 hospitals and community health centers (CHCs).  There are also 285 transit and commuter rail stops.
• A 250 m2 grid file was used for developing the coverage measure.  The proportion of the total population in each town covered in the accessibility buffer was calculated by mode for each unique activity.  Only the relevant population is considered in each step of the analysis.  For hospitals and transit, the total population is considered.  For schools, the population between 5 and 18 years of age is considered.  For higher education institutes, the population over 18 years of age is considered.  Accessibility buffers cross over town boundaries, and the accessibility of surrounding towns increases, even if the activity is not directly located within the boundary.
• The service area analysis measures accessibility based on the location of destinations.  It allows for the measurement of the access residents have to a particular activity.  Attractiveness is addressed by weighing the amenities.  Individuals have a number of choices in hospitals, while schools have equal weights.  Individuals have a number of choices for higher education.  Commuter and transit are based on frequency during the day.
• Travel times and speeds were estimated for the different modes.  Walking uses a 10 minute buffer, assuming 3 mph walking speed, while avoiding highways.  Biking uses a 10 minute buffer, assuming 10 mph biking speed, while avoiding highways.  Driving uses a 10 minute buffer and incorporating various speed limits based on the road classification.
• The Weighted Cumulative Coverage was calculated using the following formula:
• The overall rank was determined using the following formula:
• The weighted cumulative coverage and the overall rank were calculated for cities in the state. Three cities – Boston, Weston, and Pittsfield were compared initially. The results indicated that Boston has an overall rank of 2.79 for driving to destinations, compared to .83 for Weston and .37 for Pittsfield. Boston had an overall rank of 3.46 for walking to destinations, compared to .21 for Weston and .32 for Pittsfield. Boston had an overall rank of 3.1 for bicycling to destinations, compared to .51 for Weston and .39 for Pittsfield.
• Next steps in the study include expanding the approach to all of Massachusetts to establish a state average and to compare all municipalities.  Another step is to account for an overlay of activities, such as schools and hospitals that are both accessible by the same population.  Adding transit as a mode in addition to destination represents another future activity.  Establishing alternate and additional measures of attractiveness is still another future step.

[ Top ]

The State of TexasTexas Homeland Security Statewide Search (TRAIL)Statutorily Required ReportsTTI Rules and Procedures

Copyright Statement and Legal Notices

Please send comments, suggestions, and queries about this web site to the UTCM Webmaster.

University Transportation Center for Mobility™
Texas Transportation Institute · Texas A&M University System · 3135 TAMU · College Station, Texas 77843-3135
ph. (979) 845-2538 · fax (979) 845-9761