Traveling Smart: Increasing Transit Ridership by Automatic Collection (TRAC) of Individual Travel Behavior Data and Personalized Feedback

Project Title: Traveling Smart: Increasing Transit Ridership by Automatic Collection (TRAC) of Individual Travel Behavior Data and Personalized Feedback

The project objectives are to (1) determine the capabilities of existing computer and communication technologies in tracking person movements across modes (car, bike, bus, etc.) and over extended time periods (e.g., week versus daily) and (2) design, develop and test a prototype application to automate the collection of travel behavior data and provide personalized advice to increase use of transit and other alternatives to driving alone.

The ability to collect and expeditiously process and personalize the travel behavior data has been a major limiting factor using activity diary approaches to travel blending. The use of global positioning systems (GPS) have been used to track buses and other vehicles but have not been applied to track the behaviors of individual riders, nonmotorized users, and their households.

In a few short years since an earlier CUTR/FDOT project demonstrated the effect such personalized advice had on travel behavior, technology has evolved rapidly to provide a promising method to decrease the costs of collecting and analyzing the data as well as tracking person movements, not only vehicle movements. In particular, the potential now exists to use personal digital assistants (PDAs) and add-on GPS modules rather than paper diaries to track person movements, not simply vehicle movements. The GPS unit provides a means of tracking time, route, speed, while the PDA provides a means of tracking items such as mode, occupancy, and trip purpose. Unlike other GPS-tracking projects, this PDA/GPS combination should give us the ability to track across all modes (bus, bike, etc.) while minimizing some of the data entry required by the individual. One added benefit is that easing the burden on the individual may allow us to track them for longer periods to truly capture variations in habits over a week, month, etc.

Our project would investigate the challenges associated with equipping/tracking people (not vehicles) while providing feedback to the traveler and route planning information to the transit agency. These problems may include loss of signal, battery power, forgetfulness to turn it on, etc. The development of the feedback system will provide the ability to provide feedback via the web or perhaps instantly.

Conduct a literature review of personal digital assistants (PDAs) and global positioning systems (GPS) applications related to travel behavior studies. From this review, identify potential market segments (e.g., transit riders, auto drivers, bicyclists, etc.), user requirements for each segment, technology constraints related to those segments and their operating environments, and performance measures.

Investigate existing technology (PDA’s, GPS units, etc.) and communication options including wireless. Review interface options available in PDAs. Identify existing databases (GIS, transit stops, bikepaths, road construction, etc.) that are readily available for the testing area (Tampa) and easily accessible to facilitate feedback on alternatives to driving alone. Conduct technical and economical evaluation of the systems and options included in Task 1. Select the type and quantity of equipment to buy.

Deliverable: Technical memorandum summarizing the literature review (Task 1) and the technology assessment (Task 2).

The purpose of this task is to develop prototype application to collect data on personal travel behavior. This task will develop simulation models to characterize various market segments (e.g., transit rider, auto driver, bicyclist). The task will require the team to design, implement and test a portable PDA/GPS activity diary system. The activity diary will collect information such as travel purpose, origin, destination, travel time, speed, occupancy, etc. Subtasks under this task include designing the service (including user devices and communication options), identifying the data to be captured, transmitted, and processed, designing user interfaces, creating the database design, and preparing system testing procedures (communication system and PDA). As part of this application development task, the team will develop the protocol or method for conducting the field test of using the prototype.

In addition to enhancing the activity diary application through the combination of PDA and GPS units, another objective of this project is to apply this tool (TRAC-IT) to help reduce total vehicle trips and vehicle miles of travel by encouraging the use of trip-chaining and substitution for all types of trips. The tool will be designed to provide individualized feedback to activity diary users. A set of rules will be identified based on input from transportation professionals at CUTR that when applied to the data collected from each individual will yield advice to that individual on ways to reduce vehicle trips and encouraging use of alternatives to driving alone. The form and frequency for providing this feedback will be an important consideration during this task.

Another important issue to be addressed during this task is whether the set of expert rules should reside on the PDA or reside in the server with the database. The advantages and disadvantages of the two approaches will be considered together with any inherent constraints that the PDA may have (i.e. memory, availability of expert software tools, battery life).

The field test will be designed to collect up to 500 person-days of data, including data from individuals who use transit and/or bike. The field test design will address the acceptance of the device and the effectiveness of the advice.

The test to determine acceptance of the device itself will depend on the perceived utility to the individual, ease of use, and portability. This may increase the quantity and improve the quality of data desired by a transportation agency.

The test to determine the effectiveness of the advice to affect travel behavior may depend on the quantity and quality of advice given, the number of PDA/GPS units available for the test, the ability to locate and recruit individuals within targeted groups (e.g., households with a transit user) and the time period for collecting data (e.g., 1 week vs. 2 weeks).

Deliverable: A description of the field test will be developed after the completion of Task 4 and will be included in the quarterly progress report for FDOT review and comment.

Conduct limited field test of the application following the protocol outlined in Task 5. Assess the application’s effectiveness in meeting user needs and other performance measures identified in Task 1

Prepare draft final report and submit to FDOT and peer panel for review and comment. Make revisions and print the contractually required number of final reports.

Prepare streaming video presentation for the NCTR website and promote the results to the public transportation industry through presentations at appropriate conferences. The results of this research will be applicable to the public transportation community as well as those transportation professionals involved in travel demand forecasting, including tracking origin-destination patterns and measuring travel activity.

Graduate students will play a role in the research and data collection as well and the analysis and report writing. Other anticipated student benefits will include synthesis of information and technology transfer support.

Making smart travel choices require understanding the operation of the transportation system and influencing the use of public transportation. This knowledge begins with collecting data to measure or monitor travel behavior. This project builds on a previous FDOT Research Center project demonstrated that the travel habits of households provided with personalized advice did increase the use of alternatives to driving alone relative to the control group. This previous project, Reducing Vehicle Trips and Vehicle Miles Of Travel Through Customized Travel Options, offered a practical exercise that led households to reappraise their needs and rationale for travel (sometimes referred to as "travel blending"). Specific suggestions aimed at use of public transit service, bike paths, trip chaining, ridesharing and e-commerce options were provided based on specific travel patterns observed in the activity diaries. A week-long travel diary formed the basis of the personalized advice provided by CUTR.

Travel blending has been used extensively in Australia with positive increases in transit ridership. Travel blending trials conducted in Adelaide and Sydney, Australia recorded significant reductions in car use among those that persevered in the program. A study of 329 households in Adelaide saw a reduction of 17 per cent in vehicle kilometers traveled (VKT) (Tisato & Robinson 1999). Trials in Perth recorded a 17 per cent reduction in VKT, an increase in walking by 16 per cent, cycling by 91 per cent and transit by 21 per cent (Brog et al., 1999, p. 562).

The application of GPS and computer-based surveys to enhance and extend travel behavior survey methods has largely focused on tracking the vehicles (Doherty, Noel, Gosselin, Sirois and Ueno, 2001; Scholdfelder, Axhausen, Antille, and Bielaire).

CUTR will review SmartBus applications of GPS and coordinate with other technology projects ongoing at CUTR (e.g. RAPTS, NCTR). In addition, a search of the TRB Research in Progress database identified several other potentially related research projects that we should examine, including but not limited to the following:

GPS for Travel Time and Speed Monitoring
Abstract: The high level of time and effort needed to process raw GPS data and map match the data to travel facilities, manage the large amount of data collected, and analyze the resulting information seriously hinders its potential use and negates its promise.
Start date: 2003/4/14
End date: 2004/10/17
Status: Active
Source Organization: Federal Highway Administration

AVL Equipped Vehicles as Probes - Phase 3
Abstract: Past phases of this research have demonstrated in the laboratory that it is possible to estimate travel speeds as well as construct travel times through corridors using existing transit vehicles equipped with automatic vehicle locators (AVL). These results could be used to augment the existing sensor network of loops, cameras and radar. This phase will incorporate these "virtual" sensors into the existing Transportation Management System real time database so they can be used by traffic management staff in the same way that loop data is used. The system will be updated to use global positioning systems (GPS).
Start date: 2003/7
Status: Programmed
Source Organization: Washington State Department of Transportation

Implementation of Personal Digital Assistant (PDA) Devices for Superpave Field Data Collection
Abstract: The objective of this study is to evaluate PDA devices for field operations and develop PDA-based data collection, reduction, analysis and reporting procedures for Superpave activities related to HMA material quality assurance by field inspectors. This is critical in field operations to produce optimal pavement performance.
Start date: 2000/5
End date: 2003/6
Status: Active
TRB Accession Number: 809155
Source Organization: Connecticut Department of Transportation

The final report will be available on the NCTR website (www.nctr.usf.edu) in HTML and pdf formats. The results of this project will be made available at conferences in Florida such as Florida Public Transportation Association, APTA, TRB and other related. A brief presentation by the PI will be recorded and available for playback as a streaming media presentation on the NCTR website.

Peer reviewers will include a representative from the appropriate TRB Committees and/or industry groups.

1. Effective means of tracking travel behavior across modes. Previous GPS-related work relate to the movement of the transit vehicle (AVL) with links to ATIS.

2. Increased transit ridership and more use of alternatives to driving alone facilitate through personalized assistance based on actual observed behaviors, not typical or average behaviors.

Transit ridership, Advanced traveler information system, geographic information system, travel behavior.

will include faculty, students, and secretarial and other support staff who will work

directly on the project and whose costs are reflected in the direct costs of the project

* Average hourly labor rate is calculated by dividing the labor cost estimate by the total number of hours of all research faculty or students.

[1] Counting The Cost : Transport Authorities Are Investing In Sophisticated Systems Showing Where Passengers Get On And Off Vehicles. ITS International. Vol. 7, Issue no. 5

[2] Maclean, S.D. and D.J. Dailey. Wireless Internet Access to Real-time Transit Information. Transportation Research Record 1791. 2002

[3] Oh, JY; Hong, JY; Lee, SK “Conceptual Design Of Client-Side En-Route Transit Information System Utilizing 3g Mobile Phones” 9th World Congress on Intelligent Transport Systems