Long-Distance and Rural Travel Transferable Parameters for Statewide Travel Forecasting Models (2012)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

75 Many statewide models are “highway only” in the sense that the models only simulate high- way modes of transportation. Therefore, the mode choice section of the Guidebook must also include guidance on transferability and a set of transferable auto occupancy rates. The surveys available for NCHRP Project 8-84 were not necessarily conducive to developing mode choice constants or coefficients. Benchmark statistics on travel mode by distance, income, and other characteristics will also be provided. 6.1 Long-Distance and Rural Mode Choice Benchmark Statistics from Statewide Models and Other Sources Statewide Model Statistics Table 6.1 provides information on input auto occupancy rates for long-distance trips by state- wide model. Not surprisingly, auto occupancy rates are generally lowest for business trips and highest for tourist trips, the latter frequently including entire families traveling together on the same trip(s). Auto occupancy rates are clearly lower for long-distance business travelers (1.33 to 1.86) than other types of long-distance trip-makers (2.06 to 3.44); long-distance auto occupancy rates are also generally higher than those found in typical urban travel models, where work trips are in the neighborhood of 1.10 and other purposes rarely top 1.8. Bureau of Transportation Statistics As noted in previous chapters, the Bureau of Transportation Statistics (BTS) published find- ings in May 2006 from the 2001 NHTS on long-distance trip-making. Tables 6.2 through 6.4, focused on mode splits, are also derived from the same 2006 BTS report and depict long- distance trips by geography and mode, geographical size and mode, and income and mode, respectively. According to BTS’ analysis of 2001 NHTS long-distance travel, respondents from urban areas were more likely to use air as a travel mode (9 percent) than their rural counter- parts (3 percent) as depicted in Table 6.2. Also, the larger the metropolitan area, the less likely respondents are to make long-distance trips by auto, as evidenced by Table 6.3. Finally, as depicted in Table 6.4, respondents with higher incomes were more likely to use air transpor- tation for long-distance travel while lower-income groups were more likely to use intercity bus for long-distance travel. C h a p t e r 6 Auto Occupancy and Mode Choice Parameters

76 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models Auto Occupancy Rates By Purpose (Minutes or Miles) Business Tourist Other Average California – – – 1.34 Florida 1.10 2.60 1.85 Indiana – – – 3.06 Louisiana 1.86 3.44 2.64 2.65 Mississippi (Interstate) 1.39 2.55 2.05 2.00 Mississippi (Intrastate) 1.50 2.55 2.26 2.10 Utah 1.33 – 2.06 1.70 Virginia 1.82 2.69 2.69 1.82 Source: NCHRP 836 Task 91: Validation and Sensitivity Considerations for Statewide Models. Table 6.1. Auto occupancy rates in statewide models. Personal Vehicle Air Other Modes Urban 87.0% 9.0% 4.0% Rural 95.0% 3.0% 2.0% Source: BTS. Table 6.2. 2001 long-distance trips by geography and mode. Personal Vehicle Other Modes MSA More Than 1 Million 85.0% 15.0% MSA Less Than 1 Million 92.0% 8.0% Outside of MSA 96.0% 4.0% Source: BTS. Table 6.3. 2001 long-distance trips by geographical size and mode. Income Personal Vehicle Air Other Modes a Less Than $75,000 91.0% 5.0% 4.0% More Than $75,000 84.0% 14.0% 2.0% Source: BTS. aIncome ranges of less than $25,000 and more than $25,000 were used for other mode/bus trips. Table 6.4. 2001 long-distance trips by income and mode.

auto Occupancy and Mode Choice parameters 77 2010 Travel Survey of Residents of Canada Table 6.5 provides information on party size of long-distance trips by purpose according to the previously referenced TSRC. Not surprisingly, the majority of business trips have a party size of one person (81.16 percent) while less than 50 percent of the other three “discretionary” purposes are one-person trips. In terms of other statistics of interest from the TSRC: • Over 90 percent of long-distance trips are made via private automobile. The same is true for every trip purpose except for business trips, which are 73 percent auto trips, 14 commercial aircraft, 3 percent bus, and 2 percent train. • Forty-three percent of business trips are made by households with annual incomes greater than $100,000. This same income group represents one-fourth to one-third of long-distance travelers for other trip purposes. • The survey includes trips of shorter distance; however, 40–79 kilometers (about 25–49 miles) is the most common travel distance for all trip purposes. Business trips tend to be longer than other types, with 28 percent in the 80–159 kilometer range, 19 percent in the 160–319 kilome- ter range and 15 percent in the 320–799 kilometer range. 6.2 Analytical Approach to Estimating Long-Distance and Rural Mode Choice Parameters and Benchmarks The same analytical procedures described previously on trip generation and distribution were applied to mode choice. ATS statistics on party size and mode choice are provided in this section. Party size was summarized by trip purpose as auto occupancy was not asked of ATS respondents. Of course, it is only indicative of auto occupancy for trips taken by personal vehicle. As depicted in Table 6.6, a party size of three or more occupants was the most common for long-distance trips at more than 45 percent, followed by two-person party sizes at 33 percent. The average party size was similar for pleasure and personal business trips (slightly less than 3.5) but considerably lower for business trips, with a mean of 2.1. Trip Duration: Total – Domestic Travel (age 18+): Person-Trips with the Destination in Canada *** Column Percents *** Main Trip Purpose Standard Person-Trip Stub Variables Total Pleasure, Vacation, Holiday Visiting Friends or Relatives Business and All Conferences or Conventions Shopping and Other >>> Final Data <<< Household Party Size: 100.00% 100.00% 100.00% 100.00% 100.00% 1 Person 44.32% 40.18% 43.57% 81.16% 43.31% 2 Persons 36.56% 37.89% 36.74% 16.43% 41.25% 3 Persons 8.74% 9.48% 9.04% 1.08% 8.71% 4 Persons 7.61% 9.30% 7.58% 0.87% 5.39% 5 Persons 2.25% 2.51% 2.51% 0.43% 1.17% Source: 2010 Travel Survey of Residents of Canada. Table 6.5. Canadian residents’ long-distance trips, party size by purpose.

78 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models Mode of travel was evaluated by trip purpose as well as distance category. Table 6.7 depicts the number and percent of long-distance trips by mode and distance category. Personal vehicle is by far the most common transportation mode for long-distance trips at nearly 82 percent. Air travel is second at just under 15 percent, although for trips of 300 miles or greater, the air mode is around two-thirds as much as the frequency of personal vehicle trips. Remaining modes each represent less than 2 percent of long-distance trips. Table 6.8 depicts the number of long-distance trips by mode and trip purpose along with the percent of long-distance trips by the three trip purposes. In terms of the percentage of trips by purpose, nearly 60 percent of long-distance trips were categorized as “pleasure,” while 22 percent were “business” trips, and 19 percent were for “personal business.” Air travel represents a much greater percentage of business trips (33,370 out of 119,171 trips = 28 percent) than other types of long-distance trips (11 percent for pleasure trips and 9 percent for personal business trips). 6.3 Mode Choice: Long-Distance Auto Occupancy Rates Most statewide passenger travel demand forecasting models focus on the highway mode because most states have limited intrastate intercity air, rail, or bus travel. The ATS sample of nonhighway modes is sufficient for providing mode splits, as depicted earlier, by trip purpose and travel distance in Tables 6.6 and 6.7, but ATS data are largely insufficient for estimating logit mode choice parameters for calibration. FHWA has a research project underway, Development Business Pleasure Personal Business Total Percentage for All Trips 01 1 Person 61,639 39,225 14,246 115,110 21.49 02 2 Persons 31,779 107,389 37,764 176,932 33.04 03 3+ Persons 25,753 169,169 48,556 243,478 45.47 Total 119,171 315,783 100,566 535,520 100 Mean 2.107 3.332 3.456 3.077 Percent by Purpose 22.25 58.97 18.78 100 Source: 1995 ATS. Table 6.6. ATS annual frequency by purpose/party size, round-trip. 100-300 Miles > 300 Miles Total Percent Mode 01 Personal Vehicle 336,744 100,672 437,416 81.68 02 Air 11,275 66,816 78,091 14.58 03 Bus 7,026 2,904 9,930 1.85 04 Train 1,502 944 2,446 0.46 05 Commercial Vehicle 2,813 3,679 6,492 1.21 06 Waterborne 170 166 336 0.06 07 Other 559 250 809 0.15 Total 360,089 175,431 535,520 100.00 Percent by Distance 67.24 32.76 100 Source: 1995 ATS. Table 6.7. ATS annual frequency by distance/mode, round-trip.

auto Occupancy and Mode Choice parameters 79 of Long-Distance Multimodal Passenger Travel Modal Choice Model, that may provide cali- brated model parameters. Therefore, the key transferable parameter for the mode choice step is the auto occupancy rate, used to convert person trips to vehicle trips for highway assignment. Table 6.9 provides mean auto occupancy rates for each long-distance trip purpose. Auto occupancy rates are considerably higher for long-distance trips than those typically (less than 2.0) found in urban and regional planning models. Table 6.5 provided a breakdown of long-distance trips by each of three auto occupancy categories. Auto occupancy rates were calculated using the ATS Party Size attribute for auto trips only. 6.4 Mode Choice: Rural Auto Occupancy Rates Mode splits are not very relevant to rural trip-making because few rural communities provide fixed-route bus services. Commuter rail and express bus are not common modes for connect- ing rural households to urban work destinations. Therefore, the key transferable parameter for the rural mode choice step is the auto occupancy rate, similar to the earlier discussion on long- distance travel. Table 6.10 provides mean auto occupancy rates for each rural trip purpose, as well as statistics for urbanized areas for comparison. Auto occupancy rates in rural areas (bold font) seem to fall somewhere in the middle of urbanized groupings, generally higher than small- and medium-sized urbanized areas but lower than the largest urbanized areas. Auto occupancy rates are depicted in Appendix I for two-plus and three-plus occupant automobiles and daily, a.m., and p.m. peak periods. Source: 1995 ATS. Business Pleasure Personal Business Total 01 Personal Vehicle 81,652 268,533 87,231 437,416 02 Air 33,370 36,107 8,614 78,091 03 Bus 963 5,735 3,232 9,930 04 Train 861 1,325 260 2,446 05 Commercial Vehicle 2,144 3,197 1,151 6,492 06 Waterborne 54 251 31 336 07 Other 127 635 47 809 Total 119,171 315,783 100,566 535,520 Percent by Purpose 22.25 58.97 18.78 100 Table 6.8. ATS annual frequency by purpose/mode, round-trip. Mean Standard Deviation 01 Business 2.11 2.51 02 Pleasure 3.33 2.97 03 Personal Business 3.46 3.96 Total 3.10 3.17 Source: 1995 ATS. Table 6.9. Long-distance auto occupancy rates.

80 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models Urban Size/Population Trip Purpose Home-Based Work Home-Based Nonwork Nonhome- Based All Trips 1 million or more with subway or rail 1.06 1.77 1.68 1.54 1 million or more without subway or rail 1.11 1.78 1.69 1.63 Between 500,000 and 1 million 1.06 1.71 1.65 1.50 Between 200,000 and 500,000 1.12 1.72 1.52 1.49 Less than 200,000 1.11 1.65 1.64 1.52 Not in an urbanized area 1.11 1.69 1.67 1.54 All areas 1.10 1.72 1.66 1.55 Source: NCHRP Report 716/2009 NHTS. Table 6.10. Rural versus urban average auto occupancy rates.