The Development of a Direct Demand Non-Home Based

The Development of a Direct Demand Non-Home Based Model for Urban Rail Travel Rhett Fussell, PE –PB Americas Bill Davidson-PB Americas Joel Freedman-PB Americas Chaushie Chu-LA County Metro Robert Farley- LA County Metro TRB Planning Applications Conference May 19, 2009 Houston, TX

Presentation Overview • Purpose of the Project • Need for this Work • Existing Non-Home Based Direct Demand Model • Proposed New Model Structure • Data Collection Effort • Next Steps

Purpose of the Project • Improve/Enhance New Starts Forecast Reliability • Collect Rail Data Specific to NHB Trips • Develop New Methodology for Predicting NHB Rail Trips

Need for the Work? • NHB rail trips are relatively unknown – No real specific data collection currently • FTA Discretionary Funding to Improve New Starts • Current Model based solely on density • Trip Models not accurate predicting NHB rail – Area doesn’t currently have rail

Need For Work • “FTA will also give priority to proposals to develop and apply methods to estimate the time savings experienced by highway users that result from transit investments as well as proposals to develop and apply simplified travel forecasting approaches for Small Starts alternatives that cannot be addressed easily by the existing regional travel forecasting procedures” 1 Federal Transit Administration, Alternatives Analysis Discretionary Program: Notice, Federal Register, Vol. 56, No. 72, March 2007, pp 13980 -13981 1

Need For Work • On the ridership side, “FTA notes that CATS’ (Charlotte Area Transit System) forecast of new riders and user benefits reflects a significant share of induced non-home-based trips. This is a market that traditional travel forecasting procedures have not typically addressed; as such, the project’s travel forecasts and subsequent estimate of cost effectiveness entail some risk. ” 2 Lowrey, Michael, Charlotte, Triangle Transit Delayed, Carolina Journal, June 2004 2

Existing Model Form • Developed in 1985 –Jim Ryan(then of PB) – Re-estimated for Honolulu 2007 • Based on Basic Input – HB Alightings at each station – Employment Density around the stations • Type 4<3000 • Type 3<10, 000 Type 2<60, 000 Type 1>60, 000 • Assumes all trips would have been made in other alternative if rail wasn’t there • Benefits =(Rail COTBuild – Bus COTBaseline) *NHB tripsbuild COT=Cost of Travel

Existing Model Form • NHB tripends = A 1*Type. Ns * HBatts + A 2*Type. Ns * HBatts + …… where: HB=Home based alightings for that mode NHB trip ends = TOTAL NHB trip ends at each station S=station An = coeff for each density type/access mode (n=density type, ie=1, 2, 3, 4) Type. Ns = indicator variable, 1 if station is type N and 0 if not

Existing Model Form • Final Utility Equation: • Java program to run the model • Outputs – Station to Station Trip table – Station to Station Benefits

Proposed New Model Structure • Tour Frequency Model -Nested Logit • Currently one away from work activity location • Could be extended to multiple stops Choice 0 2+ TAZ 1 Drive 1 TAZ 2 TAZ 3 Walk Bus Rail # of Tours Destination Mode

Data Collection Effort • Use Current Home Interview Survey – Proportion of work based sub-tours downtown – Workers by Industry for expansion use • Employment Data – TAZ data- more detailed than model – Used to get station density/employment types • Station Counts – Hourly count data for our survey stations

Data Collection Effort • Midday Rail Station Survey – 6 stations – Varying access levels to rail – Varying densities/employment types – Alighting passengers interviewed – DASH circulator interviewed too

Data Collection Effort • “Minor” Workplace Survey – No employee counts needed – Web-based –incentive based – Collect midday work based travel – Transit accessibility – In downtown and outside of core

Data Collection Effort Sample Screenshots

Data Collection-Minor Workplace Survey • • • • Workplace location Employment type by industry code (NAICS and SCAG model type) Occupation/title Work status (full-time or part-time) /number of hours worked per week. Primary mode used to commute to work Transit pass ownership Parking cost (if drive to work) Number of times per week respondent leaves work between 10 am-4 pm Gender Age (categorized) Household size Number of workers in household Number of household vehicles Annual salary or annual household income (categorized)

Data Collection-Minor Workplace Survey • Did you leave work during the previous day (between 10 am-4 pm) to a destination greater than one block away? • Did you return to work before the end of the day?

Data Collection-Minor Workplace Survey • • • – – – Location and address of each destination Purpose of trip (eat, shop, work-related business, personal business (such as a medical or dental appointment), etc. ) Departure and arrival time for each destination Mode used for each destination (walk, bike, bus, rail, drive) If the mode to a particular destination is rail or bus, the respondent will be asked for additional information about the trip, such as: Rail line/route number used Access mode Access station/stop Egress mode Egress station/stop Fare paid If the mode to a particular destination is auto, then parking cost and vehicle occupancy will be asked

Data Collection-Station Survey • • • Have you been at work today? Are you returning to work before the end of the day? Origin of last place(since we are asking at the destination station) What were you doing there? (purpose on) What are you doing at this destination station? (purpose off) Final destination(address of where they are going) Actual work location? (address) Employment type of their occupation(industry type) Income level How would they make this trip if rail was not available (walk, bus, drive, not at all) How many days per week do you leave work to make trips between 10 am-4 pm?

Next Steps • • Recruit Employers for Minor Survey Do the Data Collection Clean the Data Begin to Develop the New Model structures Finalize in Phase II the New Model Apply to Other Test Areas FTA Buy-in for New Starts New Program for Running Model

Questions? Rhett Fussell, PE Parsons Brinckerhoff 400 SW Sixth Ave Portland, Oregon 97209 503. 478. 2802 fussell@pbworld. com