Trip Generation at Mixed-Use Developments December 19, 2012 Jerry - PDF document

12/27/2012 Trip Generation at Mixed-Use Developments December 19, 2012 Jerry Walters ◊ Fehr & Peers EPA MXD 1

12/27/2012 “D” Factors that Affect Trip Generation 1. Density dwellings, jobs per acre 2. Diversity mix of housing, jobs, retail 3. Design connectivity, walkability 4. Destinations regional accessibility 5. Distance to Transit rail proximity 6. Development Scale pop, jobs 7. Demographics household size, income 1D 2

12/27/2012 EPA Nationwide Study of MXD Travel 239 MXD: Seattle, Portland, Sacramento, Boston, Atlanta, Houston Validation: Northern and So. Cal, Texas, Georgia, Florida, Utah River Place, Portland Gateway Oaks, Sacramento 7D Factors Correlated with Reduced Travel • Density of population and employment • Diversity: jobs/housing relative to regional balance • Diversity: balance of commercial, office, and public • Design: intersections per square mile • Destination Accessibility: jobs within 1 mile • Destination Accessibility: jobs within a 30 min by transit • Distance to Transit: rail station, bus stops within ¼ mile • Development Scale: MXD population and employment • Demographics: household size, vehicle ownership * Internal travel and walking, transit use, trip length 3

12/27/2012 27 Nationwide Validation Sites • 6 Florida sites (including ITE Trip Generation Handbook) • 15 California sites • 2 sites in Texas • • 2 in Georgia, S Carolina • 2 sites in Utah Variety of scale, mix, design Atlantic Station, Atlanta Uptown District, San Diego 4

12/27/2012 Irvine California Plano Texas Mixed-Use Centers, California and Florida 5

12/27/2012 Otay Ranch California Celebration Florida South Davis, CA South Davis, California 6

12/27/2012 Moraga, CA Mockingbird Station, Dallas Bay Street, Emeryville, CA 7

Error Comparison: EPA MXD v. ITE Handbook External Vehicle Trips (1000s) 100 10 20 30 40 50 60 70 80 90 0 Site 1 Site 2 Site 3 Site 4 Site 5 Gross Trips Site 6 MXD Model Validation Site 7 Site 8 (27 Validation Sites) Site 9 Site 10 Net Trips Site 11 Site 12 Site 13 Site 14 MXD Model Site 15 Site 16 Site 17 Site 18 Site 19 Site 20 Observed Site 21 Site 22 Site 23 Site 24 Site 25 Site 26 Site 27 Site 28 12/27/2012 8

12/27/2012 MXD Model Compared with ITE MXD ITE 100 90 Observed Trips (1000s) 80 70 60 50 40 30 20 10 0 0 20 40 60 80 100 Predicted Trips (1000s) MXD Acceptance 9

12/27/2012 NCHRP 684 NCHRP 684 Internalization Surveys Percent Internal Capture AM Peak Period PM Peak Period Development At Trip Origin At Trip Destination At Trip Origin At Trip Destination (Outbound Trips) (Inbound Trips) (Outbound Trips) (Inbound Trips) Mockingbird 31% 22% 36% 38% Station Atlantic Station 17 12 38 44 Legacy Town 11 15 37 33 Center Country Isles -- -- 22 24 Village Commons -- -- 9 9 Boca del Mar -- -- 8 7 10

12/27/2012 NCHRP 684 Land Use Interactions 1 Destination Land Use Origin Land Use Office Retail Restaurant Residential Cinema Hotel Office -- 20% 4% 2% 0% 0% Retail 2% -- 29% 26% 4% 5% Restaurant 3% 41% -- 18% 8% 7% Residential 4% 42% 21% -- 0% 3% Cinema 2% 21% 31% 8% -- 2% Hotel 0% 16% 68% 2% 0% -- Effect of Separation on Interaction 11

12/27/2012 Error Comparison: NCHRP vs ITE Handbook (3 Validation Sites) Existing ITE Error Type NCHRP Method Explanation Method Average error +26% -4% Average error for sum of all sites Absolute average error 28% 17% Average magnitude of error per site Expect two-thirds of site estimates Standard deviation 34% 20% within this error range Integrated Method EPA MXD + NCHRP 684 12

12/27/2012 Optimal Blend of NCHRP 684 and EPA MXD AM Peak PM Peak ADT NCHRP 684 10.1% 36.5% n/a EPA MXD 89.9% 63.5% 100% The MXD+ Method for Estimating Traffic Generation of Lower-impact Development 1. Apply the full EPA MXD method to predict external traffic generation as influenced by site development scale, density, accessibility, walkability and transit availability, resident demographics and general mix of uses. 2. Apply the full NCHRP 684 method to capture the effects of detailed land use categories, including hotel and theater and restaurant, and the spatial separation of uses within small and medium sites. 3. Combine the results of the two methods in terms of percentages of trips remaining internal to the development site, using proportioning factors presented in the table above. 4. Apply walking and transit adjustments to off-site travel using the EPA MXD method. 5. Discount standard ITE traffic generation rates by the percentages of internalization produced in step 3 and the percentage of walk and transit travel in step 4 to obtain the estimate of site generated traffic. 13

12/27/2012 Comparison of Methods Development, Validation, Acceptance + EPA MXD NCHRP 684 MXD Method Method Method Household surveys at 239 sites in Traffic counts and Combined sources used Development of Boston, Atlanta, Houston, interviews at 3 sites in for EPA MXD and Method Portland, Seattle, and Sacramento Georgia and Texas NCHRP 684 7 sites, including 3 sites at which model 27 sites in CA, UT, TX, GA, FL, NC Combined sources, 27 development surveys Validation of Method including all of the NCHRP 684 sites, used for EPA MXD were performed, plus 3 survey sites and NCHRP 684 sites in Florida and one in N Carolina ASCE Peer Review, SANDAG MPO and traffic engineers, and public Acceptance in and agency review of NCHRP review panel (New method) Profession environmental documents in CA and WA. Comparison of Methods Project Characteristics Considered MXD + EPA MXD NCHRP 684 Method Method Method   Density of Development    Diversity of Uses: Jobs/Housing   Diversity of Uses: Housing/Retail   Diversity of Uses: Jobs/Services   Diversity of Uses: Entertainment, Hotel    Design: Connectivity, Walkability   Design: Separation Among Uses   Destination Accessibility by Transit   Destination Accessibility by Walk/Bike   Distance from Transit Stop Development Scale     Demographic Profile 14

12/27/2012 Comparison of Methods Performance at Validation Sites MXD + EPA MXD NCHRP 684 Method 1 Method Method Daily Traffic Generation R-squared 96% -- 96% Average Error 2% -- 2% Root Mean Square Error 17% -- 17% AM Peak Traffic Generation R-squared 97% 93% 97% Average Error 12% 30% 12% Root Mean Square Error 21% 33% 21% PM Peak Traffic Generation R-squared 95% 81% 97% Average Error 8% 18% 4% Root Mean Square Error 18% 36% 15% Daily Traffic Generation Comparison of ITE Handbook and MXD+ Methods ITE Handbook MXD+ (EPA) Traffic Count 90,000 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 15

12/27/2012 AM Peak Hour Traffic Generation Comparison of ITE Handbook and MXD+ Methods ITE Handbook MXD+ Traffic Count 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 PM Peak Hour Traffic Generation Comparison of ITE Handbook and MXD+ Methods ITE Handbook MXD+ Traffic Count 9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 16

12/27/2012 Trip Generation at Mixed-Use Developments December 19, 2012 Jerry Walters ◊ Fehr & Peers 17