1. Applications in Mobile Technology for Travel Data Collection 2012 Border to Border Transportation Conference South Padre Island, Texas November, 13, 2012 Ed Hard, TTI

2. Technology in Travel Data Collection 2 Bluetooth GPS data mining Cellular location data Smart phone ‘Apps’ enabled with GPS Combinations of above

3. Wireless technology for exchanging data over short distances Bluetooth frequently embedded in mobile phones and in-vehicle navigation systems Every Bluetooth device has a unique Media Access Control (MAC) address Bluetooth devices can be anonymously detected Commonly used in developing travel time and speed estimates, and more recently O&D data Bluetooth Overview 3

4. Bluetooth Technology 4

5. I-45 Deployment Travel Time Matches Rural incident impact on 13 mile segment

6. West Houston Deployment 14 Arterials, 160+ directional miles of roadway 2 to 8 lane arterials 50 Readers in the grid network

7. Collects samples of ‘actual’ trip-making that can be expanded to total traffic Each unit collects data in both directions Will not double/triple count vehicles with multiple Bluetooth devices Collects ample percentages of Bluetooth ‘reads’ to total traffic – 5 to 25% of traffic, depending on roadway and area – More than adequate sample sizes Key Points About TTI’s Bluetooth for O&D Data Collection 7

8. O&D Data Collection Using Bluetooth Field Test in Bryan-College Station Deployed BT readers at same sites as prior external survey Travel time runs prior to data collection Collected data for 72 hours Vehicle Classification counts at all external sites Compared BT results to prior external survey results 8

9. Bluetooth Observations 9 Station # (TAZ) Site Description Aug. 16Aug. 17Aug. 18 Total3-day Count* 3-day % Reads TueWedThu 479OSR @ Madison Co Line861021052933,3578.7% 480SH 21 @ Madison Co Line5796275771,78322,0508.1% 481Democrat Rd @ Grimes Co Line868224684.7% 482FM 2038 @ Grimes Co Line9715316215.0% 483SH 30 @ Grimes Co Line4925574771,52620,3127.5% 484SH 6 @ Grimes Co Line2,0462,1802,1406,36677,6438.2% 485FM 159 @ Washington Co Line142128631,2944.9% 486FM 60 @ Burleson Co Line5926295871,80826,0706.9% 487SH 21 @ Burleson Co Line1,0571,1671,2293,45339,0278.9% 488FM 50 @ Robertson Co Line10273872622,9408.9% 489SH 6 @ Robertson Co Line1,5631,6711,6974,93173,9866.7% 490FM 46 @ Robertson Co Line891431103425,9715.7% 491FM 1940 @ Robertson Co Line4256321302,4885.2% TOTAL6,6797,2397,09221,010276,2277.6%

10. Unexpanded Through Trips 10 Station #Site DescriptionTotal Observ.Through MatchesLocal TripsPercent Through 479OSR @ Madison Co Line293212727.2% 480SH 21 @ Madison Co Line1,7831341,6497.5% 481Democrat Rd @ Grimes Co Line221214.8% 482FM 2038 @ Grimes Co Line310 0.0% 483SH 30 @ Grimes Co Line1,526841,4425.5% 484SH 6 @ Grimes Co Line6,3667555,61111.9% 485FM 159 @ Washington Co Line632613.2% 486FM 60 @ Burleson Co Line1,808471,7612.6% 487SH 21 @ Burleson Co Line3,4532153,2386.2% 488FM 50 @ Robertson Co Line2622823410.7% 489SH 6 @ Robertson Co Line4,9317714,16015.6% 490FM 46 @ Robertson Co Line342133293.8% 491FM 1940 @ Robertson Co Line13041263.1% TOTAL21,0102,07518,9359.9%

11. Comparison of 2002 and 2011 Results Percent E-E Trips 11 E-E Trips 2002 – 8% 2011 – 10%

12. Bluetooth O-D Trip Matrix Generator 12

13. 13 Developing O-D Matrices Using GPS Data Streams TTI Research – Data mining, no equipment deployment – Establish trips by direction at external stations – Evaluate quality/quantity of trip data – O&D’s determined using trip end algorithms – Analyze for local/through movements and trip tables Private firms developing ‘O&D products’ Challenges – Validation of GPS-derived O&D data – Acquiring, aggregating GPS data from private sources

14. 14 O-D Data Using GPS Data Streams

15. 15 Cellular Locational Travel Data Cell signal data used to estimate travel patterns and flows Device location determined based on cell tower triangulation, data anonymized Device ‘home’ and ‘work’ locations determined Provides travel ‘flows’ of population movements Uses still evolving, possibly best for long distance travel Challenges: – Imprecise location estimates – Less frequent sampling rates – Estimates device locations, not necessarily trip ends

16. 16 Cellular Travel Data Home-Work Trips Source: AIRSAGE TMIP webinar, April 2012

17. 17 Smartphone ‘Apps’ to Collect Travel Data Future of travel survey data collection? Real-time GPS and trip/activity logging – Interactive or passive data collection – Prompted recall on device or via web follow-up Trip times, speeds, lengths, purpose, routes can be collected Challenges – Technical: battery life, data storage, dual functionality – Privacy, recruitment, bias – Respondent burden

18. 18 Smartphone Travel Data Apps Examples NuStat’s RouteScout PTV Pacelogger

19. Questions Ed Hard e-hard@tamu.edu (979) 845-8539