Surveying and Modeling Long Distance Trips Stacey Bricka TTI Erik Sabina DRCOG Catherine Durso University of Denver Julie Paasche PTV NuStats Presented at the 13 th National TRB Transportation Applications Conference ID: 763685
Download Presentation The PPT/PDF document "Surveying and Modeling Long Distance Tri..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Surveying and Modeling Long Distance Trips Stacey Bricka, TTI Erik Sabina, DRCOG Catherine Durso, University of DenverJulie Paasche, PTV NuStats Presented at the 13 th National TRB Transportation Applications Conference May 11, 2011 -- Session 17 --
Front Range Travel Counts Project2 4 MPO Regions NFRMPO DRCOGPPACGPACOGSurveysHouseholdCommercial VehicleExternal StationLong DistanceNFRMPO DRCOG PPACG PACOG
Why Study Long Distance Travel?Estimate flow between neighboring regional areas Support model development for 4-MPO region and/or statewideSupport current and expected multi-regional studiesUnderstand special travel marketsToll facilitiesPotential inter-regional transit 3
Research Questions4 What type of long-distance travel do we get from a LD survey as compared to a 24-hour survey? What can we learn from the DRCOG survey to inform the design of future travel surveys? For the survey DRCOG conducted, what data work is needed to incorporate it into the model?
Presentation Overview5 Methods to Capture Long Distance Travel Front Range Travel Survey Long distance travel survey designLong distance travel survey resultsDRCOG ModelIncorporating long distance travelPreliminary Conclusions
Methods Considered6 Video license plate capture Targeted sampling Supplement to 24-hour diary
Video License Plate Capture7 Three sites along I-25 Capture Both directionsSunrise to 9 amMatch platesCommutersTraveling into DenverThrough trips
Source: Alliance Transportation Group
Targeted Sampling9 Use Census Data to identify long-distance commuters Randomly sample addresses from identified tracts/block groups Screen for long distance travelers
Long Distance Survey Supplement10 Extend travel period to capture long distance travel
Long Distance Surveys1995 American Travel Survey2001 National Household Travel Survey2001/2 Ohio Long Distance Survey 2004/2009 Michigan Travel SurveysRetrospective or forecast?Length of diary period? Definition of long distance trip?Trip lengthTrip purposeData elements? 11
Front Range Long Distance Survey12 2-week retrospective If no long distance trips reported, queried about most recent qualifying trip Definition: 50 miles or more, one-wayAdministrationIntegrated into 24-hr HH SurveyMailed to households already surveyedProvided with travel log for new travelers
Front Range Long Distance Survey Region 24-Hour SurveysLD Surveys MailedLD Surveys Completed Participation RateFort Collins /Loveland1,50589421223.7%Denver7,3176,1201,71428.0%Colorado Springs2,6012,36683835.4%Pueblo98986436842.6%Total12,41210,2443,12330.5%
Survey Results Region# Surveys# Trips*# LD Trips*Avg. Distance Fort Collins /Loveland212631342246 milesDenver1,7144,3203,130405 milesColorado Springs8382,5021,866284 milesPueblo3681,227781214 milesTotal3,1238,6806,119331 miles *Per 2-week period
Survey Results Preliminary Results Unweighted Data
Survey Results
Survey Results
Survey Results
Survey Results
DRCOG ModelNew activity-based model“Portland – San Francisco - Sacramento Family” of models “Fully” disaggregate (mostly)DemographicsGeography (all HH and work locations get x/y point)“Trip distribution” through tour destination and intermediate stop discrete choice models Accurate long-distance trips are crucial to estimating such models for large areas
Incorporating Data into ModelEstablishing long distance trip rateEnhancing origin-destination matrix Questions:Do we need to weight the long-distance trips differently?What is the incidence of long distance trips?How does it differ from the incidence of trips in the 24-hour diary data?How can this data enhance the origin-destination matrix?
Reweight Data? Region HH SizeHH Workers IncomeFort Collins /Loveland(no difference)LD HH have more workers(no difference)DenverLD HH largerLD HH have more workersLD HH reported higher incomesColorado Springs(no difference)LD HH have more workersLD HH reported higher incomesPueblo(no difference)(no difference)(no difference) Reweight
Long Distance Trip Incidence?24-hour diary“Record all places visited.” If travel outside the state, record city and state 14-day diary“Record all long distance trips made by household members for the two-week period.”“A long-distance trip is a trip made to a location 50 MILES AWAY or more from your home.”
Long Distance Trip Incidence?24-HR DATA 196 trips >= 50 miles 109 HH1.79 trips per LD HH per 24-hr period 0.068 trips per HH per day14-DAY DATA3,926 trips >= 50 miles2,868 HH1.36 trips per LD HH per 14-day period 0.097 trips per HH per dayN=3,132 HH completed 14-day diaryN=2,868 HH had both 24-hr and 14-day trips
Diversity of Travel? Do we capture greater diversity of long distance travel in the LD survey? Compare O-D between 24-hour data and 14-day data6 geographies:4 MPO areas (NFRMPO, DRCOG, PPACG, PACOG)Outside the MPO areas but within ColoradoOutside Colorado
24-hour
14-day
Difference
Preliminary ConclusionsLong-distance data enhances travel modelsEstimates of long-distance trip making Provides insights into inter-regional and statewide travelBest to include long-distance survey in design from the start.Higher response ratesEasier to work with dataPending: Most recent LD trip
Thank you!Stacey Bricka – s-bricka@tamu.edu Erik Sabina – esabina@drcog.org