Travel Time Reliability in the SLOCOG Region October 27, 2014 San Luis Obispo Council of Governments Transportation Education Series 1
Transportation Education Series - Agenda Grab Some Free Lunch!! Introductions and Presentation Format Travel Time Reliability US101 Mobility Master Plan What is Travel Time Reliability? Data Collection for the Travel Time Reliability Metric Travel Time Reliability in the SLOCOG Region Conclusions and Q/A 2
Presenters Jim Damkowitch, Kittelson & Associates, Principal Planner Dr. Richard Dowling, Kittelson & Associates, Senior Principal Darryl depencier, Kittelson & Associates, Planner Jorge Aguilar, Wallace Group, Principal 3
Host: San Luis Obispo Council of Governments Sponsors: Kittelson & Associates and Wallace Group SLOCOG Member Agencies: Paso Robles Atascadero San Luis Obispo Pismo Beach Arroyo Grande Grover Beach County of San Luis Obispo Regional Partners: Caltrans District 5 San Luis Obispo Regional Transit Authority San Luis Obispo Air Pollution Control District 4
KAI - Transportation Education Series (TES) Promote professional development & advancement Wide range of topics (what s of interest) Travel Time Reliability (today) Multi-modal LOS & Highway Safety Manual (SLO July 17, 2012) KAI and Wallace Group plan to co-host another TES in the Feb- March timeframe on the topic of Emerging Trends in Planning and Design TES counts for Continuing Education Requirement credits (2.0) 5
US101 Mobility Master Plan 6
Smart Mobility Framework 7
Smart Mobility Framework 8
Smart Mobility Framework 9
US101 Corridor Mobility Master Plan 10
US 101 Corridor Mobility Master Plan & Travel Time Reliability Incorporated as part of the freeway operational analysis Travel Time Reliability first application in District 5 Travel Time Reliability represented by Buffer Time Buffer Time added increment of time required to ensure you reach your destination at the desired time 95% of the time. How much earlier do you need to leave your home to ensure arriving on time. Time you could spend at home (effect is assumed to be similar to when you experience delay while driving your car i.e., travel time delay) Buffer time reduction and delay reduction added to yield total monetary time saving benefits of US101 mainline improvement concepts Results? But first - what is travel time reliability? 11
What is Travel Time Reliability? Why is Travel Time Reliability Important? Travel Time Reliability Concepts and definitions Travel Time Reliability applications for performance measurement Travel Time Reliability analysis tools Data needs for measuring travel time reliability 12
Who are the Customers? 13
Customer Needs - Travelers 14
Customer Needs Goods Movement 15
What are the Causes of Unreliability? 16
Why is Travel Time Reliability Important? 1) Our roadway networks are more frequently operating in a near or above capacity 2) Uncertainties in travel time adversely impact us in multiple ways 3) We don t have a way to measure the benefit of many of our strategies and investments http://ops.fhwa.dot.gov/publications/tt_reliability/ttr_report.htm 17
Why is Travel Time Reliability Important? Old days (Capacity-oriented) Network build-out and expansion Secure funding environment Traditional performance metrics New way (Reliability-oriented) How best to manage the system we have Financial, environmental and public perception problems Improvements that affect reliability more than capacity 18
Why is Travel Time Reliability Important? Paradigms are shifting MAP-21 (Federal) SB 375 (CA) SB 743 (CA) 19
Why is Travel Time Reliability Important? It will be required MAP-21 FHWA Notice of Proposed Rule Making February 2015 You miss out on many cost-effective solutions unless you consider reliability. TSM&O Traffic system management and operations strategies Ramp metering, HOV lanes, Express Lanes, signal optimization Active Traffic and Demand Management strategies Proactive traffic management (anticipating breakdowns). Dynamic ramp metering, Dynamic tolls, traffic adaptive controls Speed harmonization, advanced queue warning, Traveler Info. Getting longer service lives out of expensive capacity improvements 20
Travel time reliability is: Definition The distribution of travel times that a traveler should anticipate if starting a trip at a given point at a given time and day. 21
Alternative Definitions Reliability is described by the variability in travel times Reliability is the amount of time that the system fails to perform adequately (classical system definition) Reliability is the predictability of users travel time experience Reliability is the likelihood of arriving on-time All are valid but the underlying theme is that travel times for the same trip are not consistent from day-to-day for a variety of reasons 22
Travel Time Distribution 23
Characterizing Reliability 24
Buffer Time Buffer Time 25
The TTI Statistic 26
The Percent < 45 Mph 27
Commonly Applied Reliability Measures 28
It can be. Is it hard? Imagine doing HCM analysis 1,000 times. But many tools available or under development that will make it easier (less hard). FHWA NPMRDS travel time monitoring data set for NHS University of Florida/FDOT spreadsheet SHRP2-C11 Model HCM 2010 Update 29
Measuring Reliability Tools for Reliability FHWA NPMRDS travel time monitoring data set for NHS Predicting or Estimating Reliability University of Florida/FDOT spreadsheet SHRP2-C11 Method HCM 2010 Update 30
FHWA NPMRDS The Federal Highway (FHWA) National Performance Management Research Data Set (NPMRDS) A vehicle-probe (cell phone tracking) travel time data base for the National Highway System (NHS) Data collected by HERE (Nokia) Average travel times by TMC links for every 5 minutes of every day back to October 2013. Passenger cars and trucks (FHWA vehicle classes 7 and 8) Its big. Requires GIS expertise and large database tools MS Access and Excel not big enough Its Accurate No data smoothing, No interpolation. Available to MPO s and State DOT s heretraffic.nhsdata@here.com 31
University of Florida/FDOT The UF/FDOT Reliability Spreadsheets Freeway Arterials How it works Allocate AADT to each hour of year Compute recurring congestion using HCM capacities For each hour of 24 hour day compute travel times for 24 possible scenarios combining weather, incidents, work zones Assign probabilities to each scenario Compute reliability statistics Lots and lots of assumptions built in Seasonal traffic variation, incident frequencies, weather frequencies. 32
SHRP2-C11 Method Developed for estimating freeway reliability How it works Estimate recurring delay based on peak hour speed. Estimate incident related delay based on peak hour v/c ratio. Compute average annual travel time with incidents. Compute 95th percentile travel time. Compute percent of trips below 45 mph. Sensitivities Sensitive to regular peak hour congestion User must bring in outside data to estimate impacts of ATDM strategies. 33
Under Development Available in 2 years How it works HCM Update Create thousands of scenarios combining weather, incidents, demand, work zones. Select capacity adjustments for each scenario. Use Monte Carlo to select a few hundred for HCM analysis. Run HCM analyses. Tally statistics. 34
1965-2000 HCM Methods 35
HCM 2010+ Reliability Method 36
What is Good Reliability? Nobody really knows. US Experience (Exhibit 37-1 HCM) 95% TTI on US Urban Freeways ranges from 1.09 to 3.60 Median is 1.47 95% TTI on US Urban Arterials ranges from 1.27 to 1.98 Median is 1.44 Recent TRB paper 37
References on Reliability FHWA Primer on Incorporating Reliability into the Congestion Management Process (FHWA-HOP-14-034) Available on FHWA website (soon) 38
References on Reliability (2) Highway Capacity Manual Chapters 36 and 37 (on web) 39
To Predict or Estimate Reliability Demand Variability (Seasonality) Weather frequencies Light, medium, heavy Rain, snow Incident and work zone frequencies Number of lanes blocked Duration To Measure Reliability Data Needs for Reliability Hourly speeds and volumes 24/7 for 6 months to a year FHWA NPMRDS 40
Case Study SLOCOG 101 Mobility Master Plan Reliability measured using Bluetooth detection Reliability estimated University of Florida method Reliability was predicted by applying UF method growth applied to measured baseline 41
US 101 Reliability Measured 10 detectors deployed along US 101 9 individual segments Data was collected continuously for 2 months 101 46 Paso Robles 46 41 Atascadero 58 41 Morro Bay 1 San Luis Obispo Pismo Beach Grover Beach Oceano Arroyo Grande 1 101 166 42
What is Detected? Any discoverable Bluetooth device! Cell phones, game systems, laptops, some cars themselves Unique ID number (MAC Address) for all Bluetooth devices Only partial address is stored to protect anonymity Any device seen by two or more detectors can be tracked as a trip Typically varies from 5 15 % of traffic Provides travel time / average speed / direction 43
Devices placed adjacent to the roadway mounted to existing structures when possible Detection range ~300 Roadside Detectors 44
BlueMAC Detectors 45
Detector / Transmitter Processor A/C or battery powered Solar charger / controller Connector ports Housing System Components 46
Travel Time Reporting Southbound Los Osos Valley Road to Avila Beach Road Bi-modal PM peak observed June 26 th, 2014 47
Travel Characteristics Trip Distance Expected Travel Time Number of Trips Average Speed (mph) Average Travel Time Standard Deviation 15 th Percentile 85 th Percentile 95 th Percentile 4.03 202 (3:22) 1214 48.5 299 (4:59) 146.3 200 (3:20) 509 (8:29) 583 (9:43) 48
PM Travel Time Reliability (4PM to 6PM) Sunday Monday Tuesday Wednesday Thursday Friday Saturday (95 th Percentile TT Avg TT) / Average TT (237s 215s) / 215s * 100% = 10% (247s 217s) / 217s * 100% = 14% (250s 221s) / 221s * 100% = 13% (244s 222s) / 222s * 100% = 10% (271s 229s) / 229s * 100% = 18% (320s 260s) / 260s * 100% = 23% (251s 219s) / 219s * 100% = 15% 49
Neighborhood Studies Linda Vista Neighborhood BCS National Championship Game 50
Cost Digiwest unit cost (max - $3,525 / unit) including BlueMAC detector with housing and power hookups Initial installation (if mounting locations are accessible via ladder) 1 Year Cellular services for data transmission (can also communicate via Fiber Optic or other existing communications network) 1 Year Cloud services for data storage BlueMAC website for data analytics and reporting Leasing and project specific costing available by request 51
US 101 Case Study in Reliability Objective Incorporate Travel time reliability as a metric used in the B/C analysis Buffer time (during typical weekday: Tuesday Thursday) Approach Existing Conditions Vehicle Operating Conditions Published Traffic Volumes and Counts Bluetooth Data (4 months) Incidents Incident Inventory in PeMS (12 months) CHP Reports (12 months) Weather Paso Robles Airport (12 months) San Luis Obispo Airport (12 months) Work zones Caltrans Website Caltrans PeMS (12 months) 52
US 101 Case Study in Reliability Approach Future Conditions Travel Demand SLOCOG Travel Demand Model (NCHRP-255 Adjustments for Daily Volumes) Caltrans K & D Factors for full 24-hour Distribution FDOT Procedure Expanded procedure for bi-directional results Validated to baseline TTI Incidents (same as baseline allowed FDOT Procedure to adjust based on volume) Weather (same as baseline allowed FDOT Procedure to adjust based on volume) Work zones (same as baseline allowed FDOT Procedure to adjust based on volume) FDOT Procedure (recap) Compute recurring congestion using HCM capacities For each hour of 24 hour day compute travel times for 24 possible scenarios combining weather, incidents, work zones Assign probabilities to each scenario Compute reliability statistics (buffer time, BTI, travel time, TTI etc.) Delta Method: (FDOT Future FDOT Baseline) + Empirical Baseline 53
US 101 Case Study in Reliability Annualize 24-hour Distribution Effects on Capacity Weather Incidents Work zones Recurring congestion Annualize 24-hour Distribution Effects on Speeds Weather Incidents Work zones Recurring congestion 54
US 101 Case Study in Reliability 55
US 101 Case Study in Reliability 56
US 101 Case Study Findings Travel Time Reliability on US 101 Generally good reliability corridor-wide < 8 minutes Southbound < 4 minutes Northbound Anticipated to not dramatically change in the future Weather not a significant factor Work zones not a significant factor Collision rates generally at or below statewide average for like facilities 57
US 101 Case Study Findings Travel Time Reliability on US 101 Where do reliability issues occur: Five-City Area and City of San Luis Obispo Southbound Direction Northbound Direction (Five-City Area) Correlates to where the greatest congestion is projected to occur Supports US101 Mobility Master Plan Buffer Time Increased B/C of HOV Improvement in Segment 1 by 8% Buffer Time Increased B/C of HOV Improvement in Segment 2 by 4% 58
Travel Time Reliability Metric We did it US 101 Case Study Findings Wasn t too painful Learned from our experience More than 4-months of data (April August) Would have been nice to test other reliability tools Expand analysis to weekend Consistent with MAP-21 provides greater support US101 Mobility Master Plan Recommendations Supports PSR-PDS Release by SLOCOG for examining operational improvement on SB US 101 Five-City Area 59
Questions? 60
US 101 Case Study Findings 61
Questions? 62