Data sources used in the evaluation of methods and parameters are, in some cases, unique to each vehicle category and in other cases, used to support numerous vehicle categories. The data sources used to support the evaluation of each vehicle category are described individually in this section, even though some of these data source descriptions are redundant across categories.
Every year in the United States, approximately 450,000 public school buses travel an estimated 4.3 billion miles to transport 23.5 million children to and from school and school-related activities, contributing about 0.3 percent of total VMT (National Highway Traffic and Safety Administration, 2002). Because most school buses travel on local and collector streets, the percentage of school bus VMT on those streets is much higher than the percentage of VMT on arterial roads and highways. School busing is the largest form of public transportation in the United States, with over 10 billion student rides provided annually (San Diego City Schools).
There are 23,900 school buses in the State of California. The average bus travels 13,000 miles each year (California Air Resources Board, 1995). Mileage accumulation rates (MAR) are higher for light heavy-duty buses (16,000 miles per year) than for heavy heavy-duty buses (12,500 miles per year). Contractor school buses exhibit considerably higher MARs (19,000 miles per year) than either public school buses (13,000 miles per year) or private school buses (9,200 miles per year). Thirty percent of total trip duration of school buses is spent idling. The California Air Resources Board estimated the statewide school bus population and VMT by counties as part of its air quality model.
In terms of data for analysis, School Bus Fleet magazine is one of the best sources of school bus statistics (http://schoolbusfleet.com/t_home.cfm?CFID=5754288&CFTOKEN=24981657). Each year it conducts a school district survey and a school bus contractor survey. The survey includes the number of buses on each route by size, number of students transported daily, total mileage, and total costs. However, data are collected for the top 100 school districts only.
The Airport Ground Access Planning Guide published by the Federal Highway Administration (FHWA) and Federal Aviation Administration (FAA) is the main source of data for airport shuttle fleet sizes (Federal Highway Administration, 1996). This document provides data for 28 cities, and has been designed to encourage regional and local planners to carry out site-specific analyses in a manner consistent with the planning process required for statewide and systemwide management systems. The planning process described in the Guide has been designed to maximize cooperation and collaboration between the airport authority and the state and MPOs responsible for the preparation of the congestion management system and the intermodal management system.
Smith Travel Research and tourist bureaus provided hotel room information for the 28 cities listed in the Guide. The Denver Commercial Vehicle Survey was used to obtain time-of-day distribution for shuttle services (Parsons Transportation Group, 2001). The Denver Survey defines shuttle services as trips whose purpose is to "Drop-off/pick-up" people.
The Taxi Fact Book is a good source of existing data on taxis for 270 cities in the United States (Taxicab, Limousine and Paratransit Association, 2002). For each city it provides summary data on population, total miles traveled, distance per trip, trips per taxi, passengers per taxi, and passengers per trip. These data were used to identify the fleet size, trips, and VMT for the regional estimates of taxi trips.
The National Highway Travel Survey includes 260 taxi trips made by residents in its current survey (NHTS, 2001). The survey shows that 24 percent of taxi trips are taken between 7:00 a.m. and 10:00 a.m., 26 percent are taken between 10:00 a.m. and 1:00 p.m., and the remaining trips are distributed evenly across the other hours of the day.
Hotel, airport, and attraction surveys collected by the Florida Department of Transportation contain trip records for 1,033 hotel visitors, 2,548 airport passengers, and 5,310 visitors to tourist attractions (TEI Engineers and Planners 2001). Each of these surveys shows taxi trips made by non-residents as a separate mode, resulting in 46 taxi trips to hotels, 181 taxi trips to airports, and 58 taxi trips to tourist attractions (for a total of 285 taxi trips).
Smith Travel Research and tourist bureaus provided hotel room information for the 28 cities listed in the Airport Ground Access Planning Guide. The 2000 Census was the source for demographic data on the number of workers in 134 cities. Additional cities can be obtained with additional work on the geographic boundaries of the taxi data compared to the Census data.
The National Transit Database (NTD), formerly Section 15 of the Federal Transit Act, is a valuable source of information on paratransit in the United States (National Transit Database, Federal Transit Administration). It provides data on transit agencies, modes of transport, the number of paratransit vehicles, VMT, and unlinked passenger trips. The project team obtained total population, population over the age of 60, and employment data from the U.S. Census Bureau database. Although the NTD database includes paratransit data from 315 cities, the team matched 220 cities and obtained household and employment data for further analysis.
Following an increase in the number of rental cars in the United States from 1992 to 2000, companies began to trim the size of their fleets. In 2000, there were 1,829,000 rental cars in the United States, but by 2002 the number had declined to 1,643,000 (see Figure A.1).
Figure A.1
Total Rental Car Fleet Size in the United States (1992-2002)
Correlating rental car fleet size with other variables, the project team found that hotel occupancy rates are directly related to the number of rental cars in the market. Nationwide hotel occupancy rates decreased between 2000 and 2002, as shown in Table A.1 (American Hotel and Lodging Association, 2003).
| Year | Hotel Occupancy Rate |
|---|---|
| 2000 | 63.7% |
| 2001 | 60.3% |
| 2002 | 59.0% |
Two published airport surveys include rental car data: the T.F. Green Airport survey (Rhode Island) and the Walker Field Airport (Grand Junction, Colorado) survey. The first was conducted for the Railway Industry Advisory Committee garage near the airport (T.F. Green Airport Authority, 2001). Nine rental car companies operate at the airport, among them Hertz, Budget, and Avis, which currently have 50 percent of the market share. These three companies operate 160 ready spaces inside the main parking garage and the remaining companies operate from satellite facilities along Route 1.
The Walker Field Airport Authority maintains rental car revenues by month for the last 12 years (Walker Field Airport Authority, 2003). Table A.2 shows the percentage of revenues by rental car companies. Although rental car fleet size may vary from the revenue data, these data are a general indication of rental car company market share at this airport.
| Rental Car Company | Revenue Share Percentage |
|---|---|
| Avis | 24.4% |
| Budget | 9.2% |
| Hertz | 17.2% |
| National | 25.4% |
| Thrifty | 17.3% |
| Enterprise | 6.5% |
Florida District 5 (Orlando area) conducted a regional study on tourism and commuter trips (TEI Engineers and Planners, 2001). This study revealed that about 42 percent of the trips to hotels and airports in District 5 are rental car trips. The project team analyzed 5,545 rental car logs in Orlando, Volusia, Marion, and Brevard Counties to determine the time-of-day distribution. About six percent of the rental cars are rented in the a.m. peak, 39 percent at midday, 11 percent in the p.m. peak, and 44 percent at night. Florida rental car data also show that the average rental car occupancy rate is 3.4.
Data from the California Department of Motor Vehicles (DMV), published by the California Energy Commission, includes rental car statistics (California Energy Commission, 2002). These show that about 76 percent of all rental cars are sedans, four percent are pickup trucks, nine percent are sport-utility vehicles (SUV), eight percent are vans, and three percent are medium and heavy vehicles.
Based on the data from the California DMV, Hertz Rent-A-Car, T.F. Green Airport, and Walker Field Airport, the project team estimated the total fleet size and annual average mileage of rental cars in 13 cities, as shown in Table A.3.
| City Name | Fleet Size | Average Mileage per Car per Year |
|---|---|---|
| San Francisco | 89,805 | 13,715 |
| Los Angeles | 88,219 | 15,054 |
| San Diego | 12,107 | 13,929 |
| Denver | 20,864 | 10,790 |
| Orlando | 20,356 | 12,051 |
| Atlanta | 19,124 | 12,926 |
| Houston | 16,448 | 13,368 |
| Detroit | 11,570 | 15,099 |
| Portland, Oregon | 8,241 | 13,960 |
| Greensboro | 2,734 | 14,540 |
| Portland, Maine | 2,194 | 13,010 |
| Sacramento | 9,913 | 13,495 |
| Houston | 8,291 | 14,039 |
The USPS is the main source for data on the public section of the package, product, and mail delivery category, and has provided data for seven cities: Atlanta, Denver, Detroit, Greensboro, Houston, Orlando, Portland. The average vehicle miles traveled per vehicle is 18.3 for light vehicles, 21.2 for medium-duty vehicles, and 103.7 for heavy-duty vehicles (most of the daily mileage for heavy-duty vehicles occurs in intercity travel at night). The average miles per day per vehicle for all these vehicles is 19.9.
Light-duty postal vehicles travel the fewest miles per day on average, but make up the majority of the total VMT. This is because light-duty vehicles constitute 95 percent of the USPS fleet. Heavy-duty vehicles have average mileages over five times greater than light-duty vehicles. Most likely, they make trips between distribution centers while light-duty vehicles are used on local delivery routes.
Data for private mail carriers are available through commercial vehicle surveys, although these do not specifically define package, product, and mail delivery vehicles or trips. The following four definitions are from the four commercial vehicle surveys reviewed by the project team:
Data from larger private mail carriers would be very insightful and should be sought for future studies.
The Quick Response Freight Manual provides a basis for estimating Urban Freight commercial vehicles. The Manual identifies quick response parameters to predict all commercial vehicles, but this definition is not intended to include commercial vehicles carrying people. The Manual outlines procedures for producing trip tables that can be assigned to highway networks for three classes of commercial vehicles:
Commercial vehicle surveys are another source of data for urban freight estimation. Truck surveys also may provide useful data, but do not provide a complete estimate of urban freight, as urban freight is not completely transported by "trucks."
Construction transport data are available primarily through commercial vehicle surveys, although these vehicle surveys do not specifically define construction transport vehicles or trips. The following definitions pertain to the four commercial vehicle surveys consulted by the project team:
The California DMV data provides information on registered vehicle types for the cities of San Francisco, Los Angles, San Diego, and Sacramento. From this data construction vehicle fleet sizes can be estimated.
The primary source of data on safety vehicles is the vehicle registration database maintained by every state government. The project team compiled data for four cities in California (San Diego, Sacramento, San Francisco, and Los Angeles) from data processed by the California Energy Commission (California Energy Commission, 2002). In this dataset, the following vehicle types were included in the safety services category:
Commercial vehicle surveys are another source of data for safety vehicles. Of the four commercial vehicle surveys reviewed by the project team, only one (Detroit) contained data on safety vehicles (Wilbur Smith Associates, 1999). In Detroit, safety vehicles included the following vehicles based on their use:
The primary difference between the vehicle registration data and the commercial vehicle survey data is that the latter does not include government-related safety vehicles (fire trucks, ambulances, and police cars).
Demographic data for the analysis of safety vehicles, including total population and total employment for each city, was derived from the 2000 Census. Government employment also was tested as a variable and was derived from Woods and Poole data (Woods and Poole).
The primary source of data on utility vehicles is the vehicle registration database maintained by every state government. The project team compiled data for four cities in California (San Diego, Sacramento, San Francisco, and Los Angeles) from data processed by the California Energy Commission (California Energy Commission, 2002). In this dataset, the following vehicle types were included in the utility services category:
Commercial vehicle surveys are another source of data for utility vehicles. Of the four commercial vehicle surveys reviewed, Atlanta, Detroit, and the Triad cities surveys contained data on utility vehicles. These surveys classify utility vehicles based on land use or industry (utility), cargo (electrical or tools/maintenance for office/government services), and purpose (maintenance). They do not include government-related utility vehicles, which represent 24 percent of total utility vehicles in the vehicle registration database.
Demographic data for the analysis of utility vehicles, including total population and total employment for each city, were derived from the 2000 Census. Government employment also was tested as a variable and was derived from Woods and Poole data (Woods and Poole, 2003).
The primary source of data on public service vehicles is the vehicle registration database maintained by some states. The project team compiled data for four cities in California (San Diego, Sacramento, San Francisco, and Los Angeles) from data processed by the California Energy Commission. In this dataset, the following vehicle types were included in the public services category:
Demographic data for the analysis of public service vehicles, including total population and total employment for each city, were derived from the 2000 Census. Government employment also was tested as a variable and was derived from the Woods and Poole data.
Commercial vehicle surveys and DMV data are reliable sources of data for business and personal services estimates. In this study, data were used from commercial vehicle surveys for Atlanta (NuStats, 1996), Denver (Parsons Transportation Group, 2001) and Detroit (Wilbur Smith, 1999) and California DMV data for Los Angeles, San Francisco, San Diego, and Sacramento. Census Bureau data were used to obtain population and employment information for those urban areas. Land use data also can be obtained from the Census Bureau to determine the magnitude and distribution of business and personal service vehicles. VIUS is another good source of data for business and personal services vehicles.
The following four definitions are from the four surveys reviewed by the project team:
