Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

17TCS Using novel data sources to support transportation planning and analysis

565 views

Published on

Instructors: Garth Appanaitis, DKS Shaun Quayle, Washington County Michael Mauch, Iteris Brian Riordan, Strava Sal Akhter, Streetlight Liming Wang and Kristin Tufte, Portland State University Chris Wright, Tara Weidner, Josh Roll, Julie Kentosh, Rich Arnold and Alex Bettinardi, Oregon Department of Transportation

Published in: Education
  • What if you had a printing press that could spit out hundred dollar bills on demand? Do you think that would change your life? ♣♣♣ http://t.cn/AisJWzdm
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Got a new Iphone 6 in just 7 days completing surveys and offers! Now I'm just a few days away from completing and receiving my samsung tablet! Highly recommended! Definitely the best survey site out there! ◆◆◆ https://bit.ly/2Ruzr8s
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Positions Available Now! We currently have several openings for social media workers. ▲▲▲ http://ishbv.com/socialpaid/pdf
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Nothing short of a miracle! I'm writing on behalf of my husband to send you a BIG THANK YOU!! The improvement has been amazing. Peter's sleep apnea was a huge worry for both of us, and it left us both feeling tired and drowsy every morning. What you've discovered here is nothing short of a miracle. God bless you. ★★★ http://t.cn/AigiN2V1
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • We called it "operation mind control" - as we discovered a simple mind game that makes a girl become obsessed with you. (Aand it works even if you're not her type or she's already dating someone else) Here's how we figured it out... ●●● http://scamcb.com/unlockher/pdf
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Be the first to like this

17TCS Using novel data sources to support transportation planning and analysis

  1. 1. Automated Traffic Signal Performance Metrics September 12, 2017 Julie Kentosh, P.E., PTOE ODOT Signal Operations Engineer 1
  2. 2. Signal Performance Metrics  Controller Enumerations 2
  3. 3. System Requirements 3  ATC Controller  Detection  Communication
  4. 4. Signal Performance Metrics  Controller Enumerations 4
  5. 5. Signal Performance Metrics 5
  6. 6. 6 Automated Traffic Signal Performance Metrics
  7. 7. 7 Automated Traffic Signal Performance Metrics
  8. 8. 8
  9. 9. Live Demo of UDOT’s System 9  ODOT’s system is currently behind a firewall http://udottraffic.utah.gov/atspm
  10. 10. ODOT’s System 10
  11. 11. Approach Volumes 11
  12. 12. Approach Volumes 12
  13. 13. Approach Volumes 13
  14. 14. Approach Volumes 14
  15. 15. Approach Volumes 15
  16. 16. Approach Delay 16
  17. 17. Purdue Coordination Diagram 17
  18. 18. Purdue Coordination Diagram 18
  19. 19. Purdue Phase Termination 19
  20. 20. Split Monitor 20
  21. 21. Split Monitor 21
  22. 22. 22 Split Monitor
  23. 23. Split Monitor 23
  24. 24. Pedestrian Delay 24
  25. 25. Pedestrian Delay 25
  26. 26. Preemption Details 26
  27. 27. Preemption Details 27
  28. 28. Arrivals on Red 28
  29. 29. Want to Learn More?  http://udottraffic.utah.gov/atspm  UDOT will host an ATSPM developer, installer & user webinar on Monday, September 18, 2017 at 8:00 AM to 9:30 AM Pacific Time. You can join the free webinar using the following link: https://connectdot.connectsolutions.com/atspm/ 29
  30. 30. Thank You • Julie Kentosh, P.E., PTOE – (503) 986-3576 – julie.l.kentosh@odot.state.or.us 30
  31. 31. Oregon Land Use PlaceTypes: Communicating Complex Land Use Data NITC Transportation and Communities Summit Workshop: Novel data to support transportation planning and analysis Portland State University | September 12, 2017 Tara Weidner - Integrated Transportation Analysis Engineer, Oregon Department of Transportation Oregon Department of Land Conservation and Development
  32. 32. Dilemma • Challenging land use communication with local planners (especially in smaller urban/MPOs) – Ownership of Land Use forecast, a key travel model input -- byzantine data/numbers – Understanding of what is implied in Land Use forecast -- housing mix, supportive multi-modal infrastructure • Loose land use definitions (e.g., TOD, mixed use) • Terminology inconsistent with travel impacts found in the literature
  33. 33. Geography Block, Block Group, TAZ GIS Database, TAZ Parcel Network Links GTFS Network spec Various Variety of Land Use Data Data Type (source) • Development Patterns • Infrastructure, Accessibility Households, Population counts (Census) Firm/Employment by type (Quarterly Census on Employment & Wages) Walk/Bike Infrastructure (local) Road Miles (Department of Public Works) Transit Stops/Service Frequency (Generalized Transit Feed Spec) Dwelling Units & Buildings (Tax AssessorData)
  34. 34. PlaceTypes are defined by characteristics of the built environment, that influence travel choices and behavior, especially those that differentiate transit, walking, & biking  Defined in Meta-analysis of land-use/transport connections “Travel and the Built Environment” (Ewing, Cervero, JAPA 2010)  Used as input in the SHRP2 C16 Rapid Policy Assessment Tool “The Effect Of Smart Growth Policies On Travel Demand” (2013) Place Types reflect the Built Environment’s impact on travel Oregon Department of Land Conservation and Development  ODOT/DLCD created a data-driven definition across the state (2014)  Key Data: EPA 2010 Smart Location Database (90+Ds by US Census block groups)  Ground-truthed 2010 PlaceType maps with 8 MPOs and DLCD regional reps
  35. 35. EPA Smart Location Database (SLD) Built Form Variables Category Variables Density Gross residential density (housing units per acre) on unprotected land Gross population density (people per acre) on unprotected land Gross employment density (jobs per acre) on unprotected land Diversity of land use Jobs per housing unit Employment entropy (a measure of employment diversity) Employment and housing entropy Urban design Street intersections per square mile High-speed road network density Transit service* Aggregate transit service frequency, afternoon peak period Transit service density, afternoon peak period Distance to nearest transit stop Destination accessibility by transit* Jobs within a 45-minute transit commute Working-age population within a 45-minute transit commute Destination accessibility by car Jobs within a 45-minute drive Working-age population within a 45-minute drive Demographics Percentage of households with no car, 1 car, or 2 or more cars Percentage of workers that are low, medium, or high wage (by home and work locations) Employment Employment totals broken down by 5-tier classification scheme Employment totals broken down by 8-tier classification scheme Over 90 Built Form Variables (2010 census block group geography across the US) https://www.epa.gov/smartgrowth/smart-location-mapping *GTFS locations only
  36. 36. 1. Destination accessibility The regional jobs that are located within 5 miles of each location, relative to the jobs reachable from a highly accessible location within the region 2. Density of jobs and housing Jobs and households per unprotected acre, within ¼ mile of each location 3. Design of street network – Multi-modal street links per square mile or – Pedestrian-oriented street links per square mile 4. Diversity of land use – Jobs to household ratio within ¼ mile – Retail and service jobs to household ratio within ¼ mile 5. Distance to transit service PM peak hourly (fixed route) transit service with ¼ mile of each location 5Ds of the Built Environment
  37. 37. 5Ds Calculations (data already collected for travel models) Jobs within 5 miles of TAZ i Jobs within 5 miles of highly accessible TAZ (85th percentile) Destination Accessibility = of TAZ i TAZ Jobs + TAZ Households + (TAZ Group Quarters/2.5) TAZ Land Area (Acres, water & parks removed) Density = of TAZ i Multi-modal Street Lane Miles in TAZ i TAZ Land Area (sqft, water & parks removed) Pedestrian-oriented Street Lane Miles in TAZ i TAZ Land Area (sqft, water & parks removed) Design = of TAZ i Best Value of 2 metrics Diversity = of TAZ i TAZ Jobs TAZ Households TAZ Retail/Service Jobs TAZ Households Best Value of 2 metrics : : Sum of (PM peak hourly Transit Service Frequency* for Route X) for all Fixed Routes Transit lines X that have stops within ¼ mile of TAZ i Distance to Transit Svc = of TAZ i Jobs, Households, GQ, Land Area: Census, QCEW, or Travel Model zone data Street lane miles: All Streets Network, or Travel Model network data Transit Service Frequency: GTFS 1 2 3 4 5 Data sources:
  38. 38. 5D Variable Measure Levels Thresholds Destination Accessibility Share of regional jobs within 5 miles High Medium Low Very Low 0.95 or more 0.85 to 0.95 0.15 to 0.85 Less than 0.15 Density Jobs & households per acre within ¼ mile High Medium Low Very Low 15 or more 5 to 15 1 to 5 Less than 1 Design Multi-modal street lane miles per square mile High Medium Low Very Low 3.3 or more 2.5 to 3.0 1.3 to 3.0 Less than 1.3 Pedestrian oriented street lane miles per square mile High Medium Low Very Low 20 or more 15.6 to 20 12.2 to 15.6 Less than 12.2 Diversity Jobs to household ratio within ¼ mile High Medium Low Very Low 1:2 to 2:1 1:2 to 1:4 or 2:1 to 4:1 1:4 to 1:8 or 4:1 to 8:1 Other Retail/service jobs to households ratio within ¼ mile High Medium Low Very Low 1:4 to 10:1 1:4-1:8 or 10:1 to 20:1 1:16-1:8 or 20:1 to 40:1 Other Transit Service PM peak hourly transit service within ¼ mile High Medium Low Very Low 150 or more 20 to 150 1 to 20 Less than 1 5Ds Levels
  39. 39. BuiltEnvironment Variables Destination Accessibility Share of Regional Jobs within 5 miles (ratio) Density Jobs & Households per acre within 0.25 mile Design Multi-modal & Pedestrian- Oriented street density (links per sq mile) Diversity Jobs (total or retail- service) to household ratio, within 0.25 mile Transit Service Level PM Peak hourly transit service within 0.25 mile 5D BUILT ENVIRONMENT VARIABLES REGIONAL ROLE (Area Type) Place Type Logic BuiltEnvironment Variables Destination Accessibility Share of Regional Jobs within 5 miles (ratio) H-M-L-VL Density Jobs & Households per acre within 0.25 mile H-M-L-VL Design Multi-modal & Pedestrian- Oriented street density (links per sq mile) H-L Diversity Jobs (total or retail- service) to household ratio, within 0.25 mile H-L Transit Service Level PM Peak hourly transit service within 0.25 mile H-M-L-VL NEIGHBORHOOD CHARACTER (Development Type)
  40. 40. Place Types Regional Role (Area Type) Regional jobs accessibility More inertia driven, less influence AREA TYPE + DEVELOPMENT TYPE = PLACE TYPE Neighborhood Character (Dev. Type) How a place functions Development type is focus for land use planning
  41. 41. Rogue Valley MPO Place Types AREA TYPE + DEVELOPMENT TYPE = PlaceType (Regional Role) (Neighborhood Character) Area Type Development Type RVMPO 2010 Place Types
  42. 42. Central Lane MPO PlaceTypes AREA TYPE + DEVELOPMENT TYPE = PlaceType (Regional Role) (Neighborhood Character) Central Lane 20-minute Neighborhood Map CLMPO 2010 Place Types
  43. 43. Transit Supportive Development • High densities of jobs and housing • High diversity of land uses, with jobs and housing • Highly accessible multi-modal transportation system • Frequent transit service (multiple routes) in peak periods Mixed Use • Medium to high densities of residential and commercial uses • High diversity of land use mix, with both jobs and housing • Multimodal transportation network supported by peak period transit service Employment • Land use is dominated by commercial or industrial activities • Low diversity of land uses • Jobs/Housing balance: mostly jobs • Missing either the density or street design required of mixed use Residential • Land use is dominated by housing • Low diversity of land uses • Jobs/Housing balance: mostly housing • Missing either the density or street design required of mixed use Rural/ Low Density • Very low densities of housing and jobs • Very low accessibility to jobs and services • Generally outside of UGB, or undeveloped areas within UGB • Auto dependent transportation, due to low activity densities Development Type Neighborhood Character Development Type is used to describe more detailed physical characteristics of each neighborhood. The Development Type of a neighborhood is determined by the activity Density, street Design, land use Diversity, and presence of transit service (level of service).
  44. 44. Communication Tool Place Type Visualization tools Web-based interactive tools invite self-paced exploring • Display zonal Place Type/5D data over Google map • Paired view (area & development types, 2 years, or 2 scenarios) • Navigation: Zoom in-out or pan • Thematic Maps: Colors indicates Place Type • Zone Pop-up: Display 5Ds values (L-M-H) & Place Types 2010 MPO Place Types Maps (arcGIS online) http://www.oregon.gov/LCD/CLIMATECHANGE/Pages/Place_Types.aspx Place Types Application Tool Examples Area Type/Dev Type side-by-side: (using json, R) http://www.oregon.gov/ODOT/Planning/Pages/PTV-SV.aspx?ptv=RVMPO-2010 Base Year/ Future Year slider: (using ESRI) http://geo.maps.arcgis.com/apps/StorytellingSwipe/index.html?appid=9e07a602cc3a41cf9ff2b61506241096# Multi-Built Form Variable Maps: (using ESRI) http://geo.maps.arcgis.com/apps/CompareAnalysis/index.html?appid=eb01b5fc6d374bd7aee59b7c68046d91
  45. 45. Placetypes_USA US-wide Census Block Group (2010EPA SLD) Interactive viewer covers 5Ds and Location, Area, Development Type at Block group level for full US: https://github.com/gregorbj/Placetypes_USA/ (R-Shiny application)
  46. 46. Communication Tool “Neighborhood Character”… Q: Why are there no “TODs” in the region? A: Transit Oriented Development (TOD), as defined in research that quantifies its impact on travel model, assumes a higher frequency of transit than found in most Oregon cities. PM Peak Transit Service Based on Google’s General Transit Feed Specification(GTFS) data used in EPA’s 2010 Smart Location Database (D4c element) at Block Group level
  47. 47. Communication Tool Opportunity Area Analysis How many more HHs do I need to reach “High” Diversity?
  48. 48. Observed Oregon Household attributes by Place Types Source: 2010 Oregon Household Activity Survey
  49. 49. Value of Oregon PlaceTypes  Better framework for thinking about land use and transportation  Criteria-driven framework for planning concepts (e.g., TOD, Mixed Use)  Engages local planners in a conversation about future land use  Visualize land use scenarios at a “neighborhood” scale  Consistency across jurisdictions allows view of comparable places  Encapsulate the complex land use relationships that drive travel  Integrate land use and transportation planning processes  Facilitate review of land use inputs to travel models  Help identify growth “opportunities” for future land use plans  Quicker development/screening of alternative land use scenarios  Better land use forecasts for models  Research/Surveys
  50. 50. Value of Oregon Place Types  Better fit than other “place type” concepts : – Neighborhood scale compatible with travel impacts – Doesn’t require extensive consultant/tool support – Localized data universally available  Oregon Place Types still evolving… Applicable to other complex datasets… – Topologies can be used to simplify complex relationship, while retaining data richness – Interactive Viewers facilitate communication/exploring
  51. 51. Ongoing Work…  Use it … 2010 Interactive maps created for all Oregon MPOs (arcGIS Online) MPOs: CAMPO, RVMPO Strategic Assessment Study , RTPs Smaller Urban Areas: Woodburn, Coos Bay, Redmond?, Pendleton? Other: RVTD Transit Master Plan, TREC Livability Survey research project  Next Steps… – Fine-Tune Method. Refine criteria to best represent observed travel behavior differences (e.g., household survey). – Existing travel modeling tools – data integration • Consistent definitions, only use data collected for travel models • Geographic scale -- Block Group vs. TAZ – Future travel modeling tools– national focus • Joint EPA Smart Location Database–National Household Travel Survey dataset, PSU research enabled direct estimation of built form variables, replacing post- processing elasticities • Land use input for national VisionEval family of Strategic Models (Placetypes_USA) • Add as land use attribute within population synthesizers? – Improve documentation/outreach
  52. 52. Resources • OSTI Scenario Planning/Place Types webpage http://www.oregon.gov/ODOT/TD/OSTI/Pages/scenario_planning.aspx#s3 • Place Types Brochure http://www.oregon.gov/ODOT/TD/TP/ORPlaceTypes/PlaceType_Flyer.pdf • Oregon Place Types Maps (arcGIS online) http://www.oregon.gov/LCD/CLIMATECHANGE/Pages/Place_Types.aspx • Oregon Place Types Online Application Examples Base Year/ Future Year change: http://geo.maps.arcgis.com/apps/StorytellingSwipe/index.html?appid=9e07a602c c3a41cf9ff2b61506241096# Multi-built form variable maps: http://geo.maps.arcgis.com/apps/CompareAnalysis/index.html?appid=eb01b5fc6 d374bd7aee59b7c68046d91
  53. 53. Questions? Tara Weidner /TPAU- Tara.J.Weidner@odot.state.or.us Brian Hurley/Planning – Brian.J.Hurley@odot.state.or.us Cody Meyer/DLCD - CMeyer@dlcd.state.or.us
  54. 54. Supporting Transportation Planning Decisions with Crash Data: ODOT Crash Data Visualization
  55. 55. Background • About me o Current Active and Sustainable Transportation Research Coordinator at ODOT o Previously Senior Planner/Modeler at Central Lane MPO • About the data o Police and self-report o Cleaned & managed by ODOT Crash Analysis and Reporting Unit o Crash, participant, vehicle o Good but not perfect
  56. 56. Objectives • Share what I know • Share what I have done • Encourage well developed data visualization, analyses, and hopefully decision making
  57. 57. Central Lane Data Portal http://www.thempo.org/887/Data-Portal
  58. 58. Data Portal Functions o Performance Monitoring o Data Access o Story Telling
  59. 59. Performance Monitoring ODOT Crash Data
  60. 60. Data Access ODOT Crash Data
  61. 61. Story Telling ODOT Crash Data ODOT Vehicle Miles Traveled PSU Population Center Data
  62. 62. Lessons • Sharing data should be important to public agencies and staff • Processed vs. raw; serve both if possible • Work the visualization into your planning effort for ease of performance monitoring • Learn a sophisticated data analysis tool (R, Python, Stata) to uncover meaningful ways to visualize the data • Tableau is good but has limitations; many competitors in the market (Qlik, Power BI, R Shiney) • Easy to port visualizations to other areas
  63. 63. Contact Josh Roll Active and Sustainable Transportation Coordinator Oregon Department of Transportation Josh.f.roll@odot.state.or.us
  64. 64. Transportation Development Division Oregon’s Experience “Using probe vehicle data to support transportation planning and analysis” Richard Arnold, P.E. Oregon Department of Transportation Transportation & Communities Summit 2017 Portland, Oregon September 12, 2017
  65. 65. Transportation Development Division Telling the Story….
  66. 66. Transportation Development Division Understanding the Data • This is vehicle probe data. It is a non- statistical sample of what is actually on the road • Collected through a variety of sources, including Mobile Devices, Portable Navigation, Commercial Fleet, Sensors, etc • Speed/Travel Time Data only (no volumes) • No vehicles => no data (usually late PM) • No vehicle => historical data
  67. 67. Transportation Development Division Average Speed for Feb 2014, on I-5 NB, between Eugene and Salem
  68. 68. Transportation Development Division SNOW STORM Feb 6, 2014 @ 114P04687 8-12 mph @ 1:30ish This looks weird!!! Reference Speed
  69. 69. Transportation Development Division What it says is all these point where we don’t have speed data (i.e., missing data), we will use the Reference Speed because it historically reflects what would be expected.
  70. 70. Transportation Development Division Different Venders, Similar Products, Different Visuals • Date providers are similar (INRIX/HERE) • There are somethings the RITIS (INRIX) does well • There are somethings that iPeMS (HERE) does well • There are somethings that neither one does best
  71. 71. Transportation Development Division Types of Applications • Corridor performance – Trends (annual, monthly, etc) – Before/after construction • Work zone analysis • Bottleneck analysis – Identify locations, categorize severity – Congestion “heat map” scans • Construction before/after studies • Reliability analysis
  72. 72. Transportation Development Division Example: Before & After • Speed Limit Changes in Central & Eastern Oregon
  73. 73. Transportation Development Division Example: Before & After • Signal Controller on US-101 through Lincoln City, OR The speed profiles for the "DISABLED" are similar to those of the "ENABLED", however, the range is greater, suggesting that travel across the "DISABLED" is less reliable; the assumption is that the on/off switch is the major factor in this different; that the weather, volumes, incidents and special events are all the same.
  74. 74. Transportation Development Division Example: Before & After • Sunrise Corridor The new alignment doesn’t exist in our data (should be available in next update), so there will not be any historical data for this either.
  75. 75. Transportation Development Division Example: Before & After • Eddyville Bypass The new alignment doesn’t exist in our data (should be available in next update), so there will not be any historical data for this either.
  76. 76. Transportation Development Division Example: Work Zone Analysis Incident Location & Time I-5 NB  TimeofDay Woodburn Wilsonville
  77. 77. Transportation Development Division Example: Solar Eclipse • Developed routes to evaluate impacts.Route ID Route Name 10 - 11 AM11 - 12 PM 12 - 1 PM 1 - 2 PM 2 - 3 PM 3 - 4 PM 4 - 5 PM 5 - 6 PM 6 - 7 PM 7 - 8 PM 8 - 9 PM 9 - 10 PM 10 - 11 PM11 - 12 AM 1222 NB US97: Bend-Redmond -2.2 -2.5 -2.6 -0.5 1.5 1.7 2.4 2.5 -0.1 0.4 -1.9 -1.7 -1.8 -2.2 1223 SB US97: Redmond-Bend -0.8 -22.1 -29.3 -21.9 -25.6 -22.8 -16.5 -14.9 -8.6 -1.9 -0.6 0 -1.9 2.1 1224 NB US97: Bend-Madras 0.1 -4.7 -2 -3.8 -3.9 -2.7 -1.7 0.3 0.1 -1.3 -3 0 -0.7 -0.5 1225 SB US97: Madras-Bend -6.1 -30.8 -35.7 -34.2 -34.6 -38.6 -36.4 -34.4 -32 -20.2 -8.5 0 -1.7 0.9 1226 EB OR126: Redmond-Prineville -3.2 2.3 0.8 -1 -0.6 0.9 -0.9 -0.8 -1.8 -1.7 -2.3 -2.8 -0.4 -0.1 1227 WB OR126: Prineville-Redmond -1.7 -2.8 0.5 -2.1 -3.9 -1.1 -2.5 -2.7 -3.2 0 -1.9 -0.1 -0.1 -1.1 1228 EB US26: Prineville-Symbiosis 0.2 -1.2 1.3 0.1 0.6 -0.8 0.1 -1.2 0.1 -0.1 -0.1 -1.7 -0.7 2 1229 WB US26: Symbiosis-Prineville -1.1 -1.9 -7 -2.6 -3.7 -14.4 -11.4 -2.5 0.2 0 -0.1 -0.3 0 0.5 1230 WB US26: Prineville-Madras -1.7 -1.9 -2.5 -1.6 -1.5 -2.4 -1.3 -0.3 2.2 0.2 -0.1 0.1 0.2 -1.1 1231 EB US26: Madras-Prineville 0.4 -8.5 -23.3 -9.5 -11.1 -4.1 -2.7 1.3 -2 0.5 -1.8 0 2.6 0 1233 WB OR126: Redmond-Sisters 1.2 -0.5 1.7 0 2 2.9 -0.9 -1.6 -0.5 -2.5 1.7 0.4 0 -0.3 1234 WB US20: Sisters-Santiam Junction -0.8 5.3 2.1 0.8 0.3 3.6 3.1 0.2 -2.2 -3.9 -2.9 -0.3 1.6 1.9 1235 EB US20: Santiam Junction-Sisters 1.9 0.5 -0.2 1 1.3 1.8 3 -0.9 0.8 1.1 0.9 0.5 -1.2 -0.2 1236 EB US26: OR216 Jct-Madras -1.7 -5.7 -4.7 -4.5 -3.4 -6.5 -4.3 -3.5 -3.3 -4.4 0.4 -0.5 0.2 -0.8 1237 WB US26: Madras-OR216 Jct -7.8 -19.4 -17.6 -18.6 -19.4 -22.6 -19.8 -19.7 -16.9 -20.7 -7.7 -0.5 0.4 0 1238 EB US26: Gumwood Ln-Madras -2.3 -16.2 -9.4 -14.3 -12.6 -19.2 -12.9 -7.5 -8.2 -7.3 -0.8 0.4 0.1 1.3 1239 WB US26: Madras-Gumwood Ln -3.6 -23 -15 -24.8 -21.5 -27.2 -29.4 -36.7 -30.8 -37.1 -5.2 -1.2 0.2 0 1240 NB US97: Madras-OR293 -2.2 -30.8 -26.3 -24.4 -18.2 -19.7 -32.8 -30.7 -22 -4.8 0.5 -1.1 -1.6 -0.4 1241 SB US97: OR293-Madras -1.4 -16.7 -35.1 -25.7 -30.4 -17.5 -11.9 -1.5 -3 -1.4 2.5 0.4 1.9 -0.9 1242 NB US97: Chestnut St-Fern Ave -2.9 -34.1 -30 -29.9 -23.3 -21 -35.8 -25.3 0.8 -4.3 -0.1 -1.5 -1.8 -0.9 1243 SB US97: Fern Ave-Chestnut St -3 -25.4 -41.9 -32.2 -38.2 -26.6 -17.7 -0.4 -5.4 -2.7 3.9 -0.1 1.5 -1 1373 NB US97: Redmond-Madras 0 -4.3 -1.4 -4.1 -2.5 -3.2 -2.8 -1.5 0.7 -2.1 -3.2 0.6 -1.5 -0.2 1374 SB US97: Madras-Redmond -8.2 -35.3 -41.6 -40.7 -42.7 -45.7 -44.5 -41.4 -39.2 -27.6 -13.1 -0.2 -1.7 0 1375 EB OR126: Sisters-Redmond -2 -1 -1.7 0.4 0.3 -0.7 -1.4 -2.7 -0.1 -1.6 -0.4 -0.7 -0.2 -0.1 1376 EB US26: Warm Springs-Madras -0.7 -11.1 -7.9 -9.8 -7.6 -12.9 -8.6 -6.3 -6.5 -5.9 0.6 0.9 0.5 -0.4 1377 WB US26: Madras-Warm Springs -6.2 -27.6 -19.4 -25.4 -24.4 -26.7 -30.5 -32.1 -29.4 -32.9 -13.3 -0.5 0.1 0 1381 EB OR22: Salem-Sisters -0.7 -0.6 -0.2 -0.6 -0.7 -0.5 0.6 -0.4 -0.3 0 0.2 -0.2 -0.4 0.8 1382 EB US20: Sweet Home-Sisters 0.7 0.1 0.3 0.5 0.5 1 1.2 -0.2 1.2 0.5 0.6 0.1 -0.4 -0.1 1383 EB OR126: Springfield-Sisters 2 3.2 2.6 2.9 3.1 2.3 3.1 1.4 1.8 1.6 -0.3 0.4 1 2.1 1384 EB US26: Sandy-Madras -2.7 -3.7 -3.4 -3 -2.4 -3.6 -2.3 -2.6 -2.9 -0.4 1.6 0 0.1 0.3 1385 EB I84: Troutdale-Madras -0.8 -2.4 -7.7 -4.9 -4.1 -0.7 -1.2 0.7 0.6 -0.5 0.8 0.7 0.6 -0.7 1386 EB I84: Troutdale-Madras -2.1 -4.2 -9.3 -3.9 -4.6 -1.6 -1.9 -0.8 -2.3 -1.1 0.5 0.7 -0.7 -0.9 1387 WB OR22: Sisters-Salem -0.2 0.5 1.5 0.6 -0.4 0.7 0.6 0.7 -1.3 -1.2 -2.1 -0.9 0.5 0.9 1388 WB US20: Sisters-Sweet Home -0.2 1.7 0.8 0.3 0.1 1 0.8 0.1 -0.6 -0.9 -0.8 -0.1 0.6 0.7 1389 WB OR126: Sisters-Springfield 4.6 5.7 7.5 6.3 7.1 6.9 6.6 6.7 5.8 5.5 2.6 4.2 3.2 3.9 1390 WB US26: Madras-Sandy -5.6 -13.7 -13.2 -17.4 -15.5 -18 -13.9 -16.5 -13 -17 -8 -7.5 -1.2 0.6 1653 WB I84: Madras-Troutdale 0.8 -1.5 4.4 9.9 15.8 17.4 -4.7 -7 -3.2 -0.9 1.2 -0.1 -0.3 -2.3 1654 WB I84: Madras-Troutdale 0.7 -3.9 3.2 8.1 9.8 13.2 -5.9 -8.7 -5.6 -1.6 -1.2 -0.7 -0.7 -2.3 Quickly identify and evaluate extent of roadway issues.
  78. 78. Transportation Development Division Example: Solar Eclipse • Average Speed on the routes
  79. 79. Transportation Development Division Example: Solar Eclipse • Tie Speed and ATR data
  80. 80. Transportation Development Division QUESTIONS??? Richard.Arnold@odot.state.or.us (503.986.4219)
  81. 81. Using Big Data for Transportation Planning & Modeling Sal Akhter sal.akhter@streetlightdata.com
  82. 82. 2-- Proprietary and Confidential -- Conventional Methods are Inadequate  Expensive  Infrequent  Time Consuming  Small Sample Size  Require Data Integration  Provide Incomplete Picture  What?  When?  Where?  Why?  How?
  83. 83. 3-- Proprietary and Confidential -- HOW  Transportation Mode  Bike/Pedestrians  Personal Vehicles  Commercial Vehicles  Medium Duty  Heavy Duty WHERE  O-D Information  By TAZs  By Zip Codes  Via Select Links  By Census Blocks WHAT  Trip Information  Speed  Volume  Distance  Duration  AADT WHY  Traveler Information  Trip Purpose  Home/Work  Demographics WHEN  Archival Information  By Month  By Day of Week  By Time of Day Leveraging Big Data to study People Mobility
  84. 84. 4-- Proprietary and Confidential -- Converting Big Data into Useful Metrics The Analysis Gap
  85. 85. 5-- Proprietary and Confidential -- Locational and Contextual Data for a Holistic View GPS Data from Vehicle & Navigation Apps 28B+ Data Points per month. Best for Understanding Trips LBS Data from Smart Phone Apps 32B+ Data Points per month Best for Understanding Activities Land Use Data to Infer Trip & Activity Purpose Census and ACS Data for Understanding Demographics Road Network Maps to Lock Trips to Routes
  86. 86. 6-- Proprietary and Confidential -- Blending Data for Accurate Information GPS Trip Data LBS Activity Data (GPS) (LBS) (LBS) (Cellular)
  87. 87. 7-- Proprietary and Confidential -- Visual Graphics and Data Download Options Data “.CSV” File
  88. 88. Case Studies
  89. 89. 9-- Proprietary and Confidential -- Pass thru trips on I-85 in Atlanta Metro area Trip Entry Trip Exit Trip Terminations Question: How many trip traveling to Atlanta metro area on I-85 NB, a major traffic corridor, are pass thru trips? Answer: The heat map shows the destinations (and relative distribution) of the trips that enter thru the zone marked “trip entry” on I-85 NB. About 3.2% of those trips continue thru the zone marked “trip exit” on I-85 NB.
  90. 90. 10-- Proprietary and Confidential -- Reducing congestion near Port of Long Beach Question: How should LA Metro re- route commercial trucks to reduce commuter congestion on I-110 near Port of Long Beach? Answer: The heat map shows the origins (and relative distribution) of all commercial trips during average weekday, peak PM hours, that use I-110 to access POLB. Destination Zone Pass thru zone at I-110
  91. 91. 11-- Proprietary and Confidential -- Construction planning at an interchange in Baltimore 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1: Early AM (12am- 7am) 2: Peak AM (7am- 9am) 3: Mid-Day (9am- 3pm) 4: Peak PM (3pm- 7pm) 5: Late PM (7pm- 12am) Share of Personal Vehicles on E. Lombard that go to each Destination Link by Day Part Bayview Blvd SB E Lombard Continuing Ponca SB Ramp to 895 NB Ramp to 895 SB Other
  92. 92. 12-- Proprietary and Confidential -- Transit routes planning in Napa county 55% 18% 12% 11% 3% Internal Trips Imported Trips Exported Trips One-way Intra-county Trips Pass-through Trips Question: Where are the transit options most needed in Napa county? Answer: Regional study showed high commuter trip volume.
  93. 93. 13-- Proprietary and Confidential -- ITS performance measurement in Ann Arbor Ellsworth Corridor where SCOOT was installed Siemens wanted empirical data to prove that its SCOOT adaptive signal technology reduced travel times. Siemens used StreetLight’s analytics to evaluate the impact. It showed SCOOT improved travel times and reliability significantly.
  94. 94. 14-- Proprietary and Confidential -- Corridor improvements impact in Toronto +87% -10% -55% -2% -3% -4% Question: Did the construction of a new highway divert traffic from local roads? Answer: A before/after study showed significant reduction of trips from neighborhood roads after the corridor improvement.
  95. 95. 15-- Proprietary and Confidential -- Transportation Demand Management in Virginia Large Volume of Short Vehicle Trips WB on Route 7 during Peak PM hours
  96. 96. 16-- Proprietary and Confidential -- Measuring a large employer’s impact on commutes Question: How can an MPO determine impact of large employers on regional commutes? Answer: A scan of greater San Diego region with 1 km2 grids shows distribution of homes for employees of a major San Diego area employer.
  97. 97. 17-- Proprietary and Confidential -- Where to add new bike facilities in Atlanta? Question: Can Big Data help planners understand where new bike facilities should be located? Answer: A scan of Atlanta region with 1 km2 grids “lit up” areas with highest volume of short trips under 2 miles.
  98. 98. 18-- Proprietary and Confidential -- Identifying popular truck routes out of POLA/POLB
  99. 99. 19-- Proprietary and Confidential -- Validating Rural AADT Counts 100 1,000 10,000 100,000 100 1,000 10,000 100,000 XDOTAADT(LOG) StL Estimated AADT (LOG) 1ST CUT - STREETLIGHT AADT ESTIMATE VS. XDOT HIGH FIDELITY LOOP COUNTERS Question: Are the DOTs old rural AADT counts accurate? Answer: StreetLight’s AADT counts returned with one outlier. Further analysis and field inspection revealed that one of the DOT’s sensor was mis-located in a much lower traffic segment of the road.
  100. 100. 20-- Proprietary and Confidential -- A Look Ahead: Multi-modal Trip Analysis
  101. 101. Thank You
  102. 102. Strava (Bike) Data Alex Bettinardi, P.E. ODOT Transportation & Communities Summit – Data Workshop September 12, 2017 Data Licensed from Strava Inc.
  103. 103. Overview • Need – very little bike / ped data • Background on Strava – And what we bought • Ways the Data has Been Used – New Bike Dataset for Oregon • Next Steps – Next Potential Purchase
  104. 104. The Need • Bike and pedestrian information is usually limited – Manual counts at intersections – Some data in travel surveys • Non-auto needs can sometimes be inadvertently over looked due to lacking information. • Lack of Data is called out a key gap in Oregon’s Bike / Ped Plan
  105. 105. Problem Statement • Without data, difficult to – Target existing investment – Illustrate benefits of past investments – Integrate bike and ped travel into planning, modeling, and decision making • Without data it is difficult to track the success of statewide efforts to improve biking and walking • Can private “big data” sources help address this need?
  106. 106. Strava – what is it? “Strava is a community of athletes from all over the world. We are a tribe. Alone or together, we strive. Strava lets you experience what we call social fitness - connecting and competing with each other via mobile and online apps. No matter the weather, day after day, we prove ourselves. Strava lets you track your rides and runs via your iPhone, Android or dedicated GPS device and helps you analyze and quantify your performance. Strava provides motivation and camaraderie, and helps us prove that we’re out there doing what we love to do.” http://www.strava.com/about
  107. 107. Strava Users • The “Prove it” subgroup (MAMILs) • A sub-sample, but still a group of valid users of the system. • Good indicators for what roads or paths get more use than others
  108. 108. Network Data – what we bought • Strava mapped bike traces to OSM (Open Streets Map) for each minute of 2013 and half of 2014 – (60 X 24 x (365+151)) = 743,040 minutes • Used OSM – needed an all-streets and all-paths base map • 310,000 of 700,000 OSM lines (edges) had at least one Strava bike trace during 2013/14
  109. 109. User Data – what we bought • ~20,400 Oregon “athletes” used Strava in 2013/14 (Oregon grew 3,000 in 5 months) • Logged ~540k bike trips/yr (“activities”) – 26 trips per year per “athlete” on average • ~ 5.6 Million BMT/yr (Bike Miles Traveled) – The “edges” have a total of 106 million bike “counts” – note a bike trip would be counted many times – once per edge (~200 edges per trip on average)
  110. 110. Strava Data: Hawthorne Bridge • One of the heaviest use areas is the Hawthorn Bridge at about 23,000 annual trips for 2013/14 (from Strava)
  111. 111. Bike Count - Hawthorne • 1.7 M WB bike counts in 2013/14 • Strava is ~1.5% at this location • Rough extrapolation ~400 M BMT Statewide – Probably between (100-600M) http://portland-hawthorne-bridge.visio-tools.com/
  112. 112. Eugene Bike Count Comparison http://www.thempo.org/documents/clmpo2013countreport.pdf
  113. 113. Almost Too Good Month Bike Counts Strava Trips Strava Trips % of Bike Counts January 10,244 89 0.9% February 13,033 125 1.0% March 17,569 202 1.1% April 20,476 205 1.0% May 23,995 230 1.0% June 24,422 239 1.0% July 28,835 308 1.1% August 16,911 166 1.0% September 146 NA NA October 15,166 160 1.1% November 12,624 125 1.0% December 7,965 60 0.8% Average 17,433 174.3 1.0% Fern Ridge Bicycle Counts and Strava Trips Comparison - 2013
  114. 114. Uses to Date - Summary • 48 request (a little over one per month) • ~10 of which for formal projects – About one a quarter – Cases where no other data really existed • Informally Strava’s free heat map has helped in many additional cases: http://labs.strava.com/heatmap/#15/- 123.02962/44.93994/blue/bike
  115. 115. Some Initial Uses – Count Locations • Helped to validate some bike count locations on the coast • Able to provide some evidence that the picked count locations were some of the highest on the coast
  116. 116. Inform Rumble Strip Conversation
  117. 117. Understanding Current Conditions
  118. 118. Evaluation Scenic Bikeways - For Travel Oregon
  119. 119. Providing Information for Bike Path Plans / Designs
  120. 120. Seasonality for Maintenance
  121. 121. Bike Warning Systems
  122. 122. Seeing where the path isn’t followed
  123. 123. Next Steps with this Data • Get 2014 information into TransGIS • Get the word out to engineers and planners that this information is available – Identify more pilot studies and locations • Continue to explore the trends – Time-of-day – Neighborhood connections – Working with MPOs
  124. 124. New Type of Data for the Agency • Strava’s product is very similar to Inrix and other speed (reliability) data products (also signal data) • Transportation data is moving towards continuous historic or continuous “live” • The interface is spatial but the data is temporal – need to develop more internal tools to deal with this
  125. 125. Unknowns Related to Bike App Data • Increasingly crowded app market • App use inherently volatile • Who is the Data Owner at ODOT? • Who is the Data Champion? • Which roles within ODOT need to have Bike considerations (and all modes) in their job descriptions.
  126. 126. Next Steps: Strava + ODOT 2.0 • Creating business case for additional purchase of 2014-17 data • Potential Research using counts & Strava to estimate bike utilization • Agency needs to continue to work with the information from these apps and other sources to fill the information gap about bike travel.
  127. 127. Questions? http://www.sheilahanlon.com
  128. 128. Using Big Data for Transportation Planning & Modeling Sal Akhter sal.akhter@streetlightdata.com
  129. 129. 2-- Proprietary and Confidential -- Conventional Methods are Insufficient  Expensive  Infrequent  Time Consuming  Small Sample Size  Require Data Integration  Provide Incomplete Picture  What?  When?  Where?  Why?  How?
  130. 130. 3-- Proprietary and Confidential -- HOW ➢ Transportation Mode  Bike/Pedestrians  Personal Vehicles  Commercial Vehicles  Medium Duty  Heavy Duty WHERE ➢ O-D Information  By TAZs  By Zip Codes  Via Select Links  By Census Blocks WHAT ➢ Trip Information  Speed  Volume  Distance  Duration  AADT WHY ➢ Traveler Information  Trip Purpose  Home/Work  Demographics WHEN ➢ Archival Information  By Month  By Day of Week  By Time of Day Leveraging Big Data to study People Mobility
  131. 131. 4-- Proprietary and Confidential -- Converting Big Data into Useful Metrics The Analysis Gap
  132. 132. 5-- Proprietary and Confidential -- Locational and Contextual Data for a Holistic View GPS Data from Vehicle & Navigation Apps 28B+ Data Points per month. Best for Understanding Trips LBS Data from Smart Phone Apps 32B+ Data Points per month Best for Understanding Activities Land Use Data to Infer Trip & Activity Purpose Census and ACS Data for Understanding Demographics Road Network Maps to Lock Trips to Routes
  133. 133. 6-- Proprietary and Confidential -- Blending Data for Accurate Information GPS Trip Data LBS Activity Data (GPS) (LBS) (LBS) (Cellular)
  134. 134. 7-- Proprietary and Confidential -- Visual Graphics and Data Download Options Data “.CSV” File
  135. 135. Case Studies
  136. 136. 9-- Proprietary and Confidential -- Pass thru trips on I-85 in Atlanta Metro area Trip Entry Trip Exit Trip Terminations Question: How many trip traveling to Atlanta metro area on I-85 NB, a major traffic corridor, are pass thru trips? Answer: The heat map shows the destinations (and relative distribution) of the trips that enter thru the zone marked “trip entry” on I-85 NB. About 3.2% of those trips continue thru the zone marked “trip exit” on I-85 NB.
  137. 137. 10-- Proprietary and Confidential -- Reducing congestion near Port of Long Beach Question: How should LA Metro re- route commercial trucks to reduce commuter congestion on I-110 near Port of Long Beach? Answer: The heat map shows the origins (and relative distribution) of all commercial trips during average weekday, peak PM hours, that use I-110 to access POLB. Destination Zone Pass thru zone at I-110
  138. 138. 11-- Proprietary and Confidential -- Construction planning at an interchange in Baltimore 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1: Early AM (12am- 7am) 2: Peak AM (7am- 9am) 3: Mid-Day (9am- 3pm) 4: Peak PM (3pm- 7pm) 5: Late PM (7pm- 12am) Share of Personal Vehicles on E. Lombard that go to each Destination Link by Day Part Bayview Blvd SB E Lombard Continuing Ponca SB Ramp to 895 NB Ramp to 895 SB Other
  139. 139. 12-- Proprietary and Confidential -- Transit routes planning in Napa county 55% 18% 12% 11% 3% Internal Trips Imported Trips Exported Trips One-way Intra-county Trips Pass-through Trips Question: Where are the transit options most needed in Napa county? Answer: Regional study showed high commuter trip volume.
  140. 140. 13-- Proprietary and Confidential -- ITS performance measurement in Ann Arbor Ellsworth Corridor where SCOOT was installed Siemens wanted empirical data to prove that its SCOOT adaptive signal technology reduced travel times. Siemens used StreetLight’s analytics to evaluate the impact. It showed SCOOT improved travel times and reliability significantly.
  141. 141. 14-- Proprietary and Confidential -- Corridor improvements impact in Toronto +87% -10% -55% -2% -3% -4% Question: Did the construction of a new highway divert traffic from local roads? Answer: A before/after study showed significant reduction of trips from neighborhood roads after the corridor improvement.
  142. 142. 15-- Proprietary and Confidential -- Transportation Demand Management in Virginia Large Volume of Short Vehicle Trips WB on Route 7 during Peak PM hours
  143. 143. 16-- Proprietary and Confidential -- Measuring a large employer’s impact on commutes Question: How can an MPO determine impact of large employers on regional commutes? Answer: A scan of greater San Diego region with 1 km2 grids shows distribution of homes for employees of a major San Diego area employer.
  144. 144. 17-- Proprietary and Confidential -- Where to add new bike facilities in Atlanta? Question: Can Big Data help planners understand where new bike facilities should be located? Answer: A scan of Atlanta region with 1 km2 grids “lit up” areas with highest volume of short trips under 2 miles.
  145. 145. 18-- Proprietary and Confidential -- Identifying popular truck routes out of POLA/POLB
  146. 146. 19-- Proprietary and Confidential -- Validating Rural AADT Counts 100 1,000 10,000 100,000 100 1,000 10,000 100,000 XDOTAADT(LOG) StL Estimated AADT (LOG) 1ST CUT - STREETLIGHT AADT ESTIMATE VS. XDOT HIGH FIDELITY LOOP COUNTERS Question: Are the DOTs old rural AADT counts accurate? Answer: StreetLight’s AADT counts returned with one outlier. Further analysis and field inspection revealed that one of the DOT’s sensor was mis-located in a much lower traffic segment of the road.
  147. 147. 20-- Proprietary and Confidential -- A Look Ahead: Multi-modal Trip Analysis
  148. 148. Thank You
  149. 149. TCS 2017 Data Workshop Highlights/Pointers of Recent NITC Research Projects on Data Liming Wang Portland State University
  150. 150. Projects Bike-Ped Portal All-in-one Spatial Activity Processer (ASAP) Vehicle Wheel Motion (VWM) and Automatic Vehicle Location (AVL) data Data integration for integrated models Ongoing projects
  151. 151. Bike-Ped Portal A national archive for bicycle and pedestrian count data. Check bicycle and pedestrian traffic volumes around the U.S. and add your own count data. Over 4 million count records from five states Researchers: Krista Nordback, Kristin Tufte URLs: http://bp.its.pdx.edu/ http://trec.pdx.edu/research/project/817 Bike-Ped Portal Phase II Ongoing Improve Bike-Ped Portal usability for both data providers and data users Researchers: Hau Hagedorn, Kristin Tufte, Nathan McNeil Webinar (forthcoming on 12/19): http://nitc.trec.pdx.edu/events/professional-development/webinar- bike-ped-portal
  152. 152. All-in-one Spatial Activity Processer (ASAP) Simplifies and standardizes GPS and related spatial activity data processing; Detects trips, travel modes and route activities from GPS traces and accelerometer data Researchers: Joe Broach, Jennifer Dill URL: Demo: http://web.pdx.edu/~jbroach/asap_demo/ Project page
  153. 153. VWM and Bus AVL data Uses Vehicle wheel motion (VWM) detection data and TriMet’s automatic vehicle location (AVL) data to quantifiy congestion and produce performance measures for arterials at the regional scale Researchers: Miguel Figliozzi, Robert Bertini Report
  154. 154. Data Integration for Integrated Models Explores methods to facilitate data integration/fusion from multiple sources for integrated transportation and land use models Researcher: Liming Wang Report
  155. 155. Ongoing projects LRT/BRT/SCT/CRT Station Area Databases, Arthur Nelson A LRT/BRT/SCT/CRT station area database for 12 light rail transit (LRT) systems, 9 bus rapid transit (BRT) systems, 4 streetcar transit (SCT), and 5 commuter rail transit (CRT) systems that is being updated to 2015. Bicycle and Pedestrian Traffic Monitoring Data Quality, Nathan McNeil and Kristin Tufte Creates a practical method to quality check bicycle and pedestrian traffic counts.
  156. 156. Bluetooth Travel Time & O/D System Shaun Quayle, P.E. Washington Co. Traffic Eng. TREC – Novel Data Sources Session September 12, 2017
  157. 157. Agenda • Bluetooth Background • County System • Use Cases – Operations – Planning
  158. 158. Need “Statement” • Realistic existing condition models  “Pass the straight face test” • Discern problem locations, days & times operationally • Proactive knowledge of issues • Alert of significant travel time change • Track travel time “rankings” over time
  159. 159. Bluetooth Background Bluetooth TM = robust, short-range wireless radio comm. protocol between electronic devices • Unlicensed, low-power 2.4 GHz band • Media Access Control (MAC) Address • Anonymous Unique Identifier (48 bit, >28 trillion) • Class 1 ~300’, Class 2 ~30’, Class 3 ~3’
  160. 160. Bluetooth Background
  161. 161. Sampling Relative to Devices
  162. 162. Washington Co. System
  163. 163. Use Case #1 – Snow Day! How does snow impact AM peak commute (7am to 9am)? Rainy Tues. Jan 10th vs. Snowy Wed., Jan 11th • Cornelius Pass Road • 185th Avenue • Murray Blvd • Durham Road • 72nd Avenue Beth Nakamura - http://www.oregonlive.com/weather/index.ssf/2017/01/january_snowfall_record-settin.html
  164. 164. Snow Day = Less Trips Trips Reduced by 75% (range = 25% to 88%) (Jan 11th snow vs. Jan 10th rain)
  165. 165. Snow Day = Faster & Slower Travel Times Average No Change (range = -40% to +48%) (Jan 11th snow vs. Jan 10th rain)
  166. 166. Snow Day = Faster & Slower Running Speeds Average -5% speed reduction (range = -32% to +68%) (Jan 11th snow vs. Jan 10th rain)
  167. 167. Use Case #2 –Signal Timing Citizen complaint(s) along Scholls Ferry Road in AM peak. Impact of signal timing adjustment, setting Nimbus/Scholls “free” Scholls Ferry Road Corridor – Bluetooth readers between 121st and Nimbus during critical AM peak.
  168. 168. INSERT RESULTS FROM EMAIL
  169. 169. BEFOREFTER Weekday 6am to 9am
  170. 170. BEFOREAFTER Weekday 7:15am to 8:15am
  171. 171. Use Case #3 –Model Calibration Need to validate existing routing.
  172. 172. OD Bluetooth v. Regional Model
  173. 173. Bluetooth Performance Measures • Travel Time & Planning Time Index • Compare 1 corridor over time • PTI = 95% TT / Free Flow TT (Reliability) • % Longer than Free Flow • Normalized, multiple corridors over time • Avg. Running Speed • Normalized, but not relatable to lay audiences
  174. 174. TV Hwy
  175. 175. Live Demo of Site
  176. 176. Summary • County has 122 roadside permanent Bluetooth readers to support • Travel time & Reliability • Running speed, and • Origin-Destination assessments. • Access to System for Agency Partners
  177. 177. Thanks to agency partners!
  178. 178. Question & Answer Shaun_quayle@co.washington.or.us ; 503-846-7938

×