In this section, the data from all of the sources discussed in the previous section are combined to provide a picture of the total magnitude and distribution of commercial vehicle travel in the 13 selected urban areas. The results have been developed for each of the 12 commercial vehicle categories defined in Section 2.0. The following measures of commercial vehicle travel have been developed:
Data also were obtained to review the magnitude and distribution of commercial vehicles by time periods and facility types, but these data were not sufficient to stratify the data by urban area or commercial vehicle category. Summary data are presented in the following sections.
After combining all of the various data sources, there were still data gaps for many urban areas. Data were obtained for all commercial vehicle categories for only two urban areas. The remaining 11 urban areas do not have data for all 12 categories. There are at least four urban areas for each individual commercial vehicle category, which provides a basis for evaluating trends across different types of cities.
Table 4.1, Table 4.2, Table 4.3, Table 4.4, and Table 4.5 present the summary statistics for total fleet size, per capita fleet size, average VMT per vehicle per day, average total VMT, and percentage of total VMT for the commercial vehicle category, respectively. These data are derived from a variety of data sources, as noted in parentheses in the table cells. While we are comparing these data across categories and across cities, it is important to recognize that the data sources may not be fully compatible, although we tried to achieve compatibility wherever possible. Although we had data from some of the individual contacts that were made, these data were not included in the summary tables because they were not comprehensive and not comparable to the other sources of data we compiled. These tables also present the minimum, maximum, and average across all urban areas for each commercial vehicle category. Finally, summaries are provided at the ends of each table on totals for the following three general categories of commercial vehicle trips: goods, people, and services. These summary statistics are described for each commercial vehicle category in the following sections.
School bus data are derived from two sources: school bus fleet data and California Department of Motor Vehicles (DMV) data. There are nine urban areas with school bus fleet data and two urban areas with data from the California DMV. As expected, the fleet sizes are generally larger for larger cities, except Atlanta, which has a much larger fleet than San Francisco and Detroit. Atlanta also has the highest VMT of all cities reviewed. The highest per capita fleet sizes occur in the smallest cities, indicating some efficiency for larger cities. There are a wide range of average daily miles per vehicle, with higher averages for smaller cities and lower averages for larger cities. For example, in Los Angeles the average VMT per vehicle is 52 miles whereas in Winston-Salem the average is 79.8 miles. Overall, the contribution to total VMT is very small (0.1 percent), but the data are not difficult to obtain or use.
There are data from five urban areas for shuttle services, derived primarily from the Airport Ground Access Planning Guide. In addition, data on shuttle services were available from the commercial vehicle survey in Denver. This category was originally intended to encompass shuttle services to a variety of destinations (airports, rail stations, bus stations, etc.) but data on non-airport sources were not readily available for evaluation. In addition, it was felt that the shuttle services to airports constituted the majority of this fleet for most cities. The statistics for per capita fleet size, percentage of total VMT, and average miles per day are very stable across urban areas. San Francisco and Orlando have the highest per capita fleet sizes, probably because of the high influence of tourism in these cities. Overall, shuttle services contribute a very small amount (0.02 percent) to the overall VMT in any urban area. In addition, the data used to characterize these services are not based on an ongoing data source and may not provide appropriate data for use over time.
All but one of the 12 urban areas (Greensboro) has data on private transportation from one data source - the Taxicab Fact Book. The per capita fleet size rates are fairly stable across urban areas, except for Houston and Orlando, which have rates that are more than double the average rate. This is again likely due to the influence of tourism in cities with a reliance on highway modes of transportation. The average miles per day are almost the same across all urban areas. Again, the overall impact on VMT is small (0.2 percent), but the data are readily available and easy to use.
All 13 urban areas have data on paratransit services, derived from the Federal Transit Administration's (FTA) Section 15 data. Los Angeles has a much higher total fleet size than any other urban area in our sample, but this is proportional to its population, and the per capita fleet size rates are similar for all urban areas. The overall impact on percent of VMT is very small (less than 0.01 percent).
The only comprehensive source of data for rental cars was the DMV data in California, where we have data for four urban areas. The California Energy Commission identified the rental cars from the master list of rental companies. These data indicate high numbers of vehicles compared to all other commercial vehicles carrying passengers. The average miles per day statistic falls between that for other passenger commercial vehicles, with shuttle service and paratransit vehicles much lower and school buses and taxis much higher. These results make intuitive sense. The per capita fleet size and percentage of total VMT is three times higher in San Francisco than any other urban area; presumably this is due to the high rate of tourism in San Francisco.
Data for package, product, and mail delivery trucks are estimated from three different sources, representing data for all 13 urban areas:
The fleet sizes for the California cities were only 10 percent of the average for this category, indicating that a majority of these vehicles were either not included in the data source or classified under another category. In the analysis of the commercial vehicle survey data, these vehicles were identified based on cargo carried and whether the purpose was for pickup or delivery, but the DMV database only captured vehicles that were identified with a body type of parcel delivery trucks. Excluding the California DMV data, the remaining urban areas have similar fleet sizes per capita, except in the Piedmont Triad cities, where the rates are very high. There is a similar trend with the percent of total VMT statistics. The average miles per day are high for all California cities, based on VIUS data, and Detroit has a similar statistic, but all other urban areas have average miles per day of less than half this value. This is assumed to be a byproduct of different definitions of vehicles in this category, as evidenced by the differing fleet sizes. For example, longer mileage per day may indicate an emphasis on product delivery where shorter mileage per day would indicate an emphasis on mail and package delivery.
There are two primary sources of data for urban freight distribution and warehouse delivery trucks: California DMV data and commercial vehicle survey data. The commercial vehicle survey data are consistently higher for per capita fleet size ratios and percent of total VMT statistics than the California DMV data. It is hypothesized that this is because we are unable to separate all business service trucks from the urban freight category in the commercial vehicle surveys, thus overestimating urban freight trucks. The combination of the urban freight truck VMT with the business service truck VMT is more consistent between the two data sources than the individual categories, supporting this theory. The average miles traveled per day is very consistent across all urban areas, with the California cities slightly higher than other cities. Overall, the California DMV data results in the shares of total VMT in the one to three percent range and the commercial vehicle survey data results in the shares of total VMT in the two to four percent range.
Again, there are two primary sources of data for construction transport: the California DMV data and commercial vehicle survey data. In this case, though, there is less variability across the urban areas with different data sources because the definition of vehicles in this category is more straightforward. The Piedmont Triad area has significantly lower fleet size and VMT than other urban areas, indicating that there may be less construction activity in this urban area (and possibly in other smaller urban areas). The average miles per day are reasonably consistent across urban areas, ranging from 31 miles per day in Denver to 58 miles per day in Atlanta. The overall share of total VMT is in the one to two percent range for all urban areas except Piedmont-Triad.
There is only one data source that provides data on safety vehicles: the California DMV data. The Detroit commercial vehicle survey data did include tow trucks, and so this data set yields data for a portion of the potential vehicles in this category, but is not comprehensive since it excludes other public safety vehicles such as police and fire trucks. The average miles per day also are lower in the Detroit data, again resulting from the different definition of vehicles in this category. The average miles per day for California urban areas in this table is based on data from the VIUS, which also only includes tow trucks. The estimate of average miles per day of police cars, derived from the individual contacts, yields a lower average of between 22 and 31 miles per day (assuming that police cars operate 365 days per year). These data are not reported in the table, because the information are estimates and not based on collected data sources.
The range of percent VMT is between 0.2 and 1.2 percent of the total VMT and the range of per capita fleet size is one to five per thousand population. Sacramento has the highest statistics in both cases, possibly because some fleets associated with the State of California may have vehicles that operate elsewhere registered in the capital City of Sacramento, similar to major companies registering fleets at the headquarters location rather than the operating location.
The California DMV data and two of the three commercial vehicle surveys contain data on utility vehicles. The commercial vehicle surveys underestimate fleet size since they include only private utility vehicles, such as trucks for plumbers and electricians, whereas the California DMV data also includes public utility vehicles, such as trash collection and meter reader vehicles. The range for the percentage of total VMT is between 0.03 and one percent, with the per capita fleet size ranging from one to four per thousand population in California cities. Again, Sacramento is quite a bit higher than the other California cities, possibly for the same reason that some public utilities vehicles may be registered in the capital city rather than in the operating city. The VIUS data for utility vehicles provides a very similar estimate for average miles per day compared to Detroit, whereas the Atlanta estimate for average miles per day is quite a bit lower. This may be a result of the different definitions in the different surveys (Atlanta defines the purpose as maintenance; Detroit has a more general utilities industry category).
The only source with information on public service vehicles is the California DMV data, which has many categories of public service vehicles at the Federal, state, city, and local government levels. In the California DMV database, government vehicles are identified strictly by type-license codes assigned by DMV on all fee-exempt records at time of registration. Per capita fleet size ranges from five to 25 per thousand population, with the highest ratio in Sacramento. The percentage of total VMT ranges from 0.6 to 3.5, also with Sacramento having the highest percentage.
The California DMV data has a large category of 'other commercial' light duty vehicles that have been allocated to this business and personal services category for our purposes. The California DMV employed the same approach used by R.L. Polk. They split and employ all key words from the 120-character owner field of each record in the database that reveal any potential business use information. Since not all of the 'other commercial' vehicles are being used for commercial purposes, this category has been factored to exclude the business and personal service vehicles being used for personal activities, based on the VIUS estimates of the use of these vehicles. In this case, personal service includes door-to-door sales and realtors and is included where personal activities are not included.
The Denver, Detroit, and Atlanta commercial vehicle surveys also include vehicles in this category, but these surveys do not include light duty vehicles, and so the estimates of fleet size are only a fraction of the actual total fleet sizes reflected in the California DMV data. This is the largest category of all commercial vehicles, ranging from three to seven percent of the total VMT for California cities. The per capita fleet size ranges from 19 to 38 per thousand population. The average miles per day are very consistent across urban areas from all sources, except in Atlanta, where it is quite a bit lower. This is likely due to the inclusion of personal activities, which are expected to be shorter duration than business and personal services.
In our original analysis, the commercial vehicle categories were aggregated into three types of vehicles, based on trip purpose. These three types were moving people, moving goods, or providing services. Table 4.6 presents a summary of fleet sizes per 1,000 population for selected urban areas by these aggregated categories. This table includes only urban areas with either a commercial vehicle survey or DMV data. At this aggregated level, the following conclusions can be drawn:
The four commercial vehicle surveys (Atlanta, Denver, Detroit, and Piedmont Triad) provide the ability to review the behavior of commercial vehicle travel by time of day. A summary of these data from the four surveys is presented in Figure 4.1. Three of the surveys show the expected distribution of trips during daylight hours, without the typical peaking that is apparent in passenger travel. The Detroit survey has a strong peak in the a.m. peak hour and relatively small numbers of trips at other times of day. This was most likely a result of the method of data collection rather than a true representation of the temporal distribution of commercial vehicle trips.
Figure 4.1: Percent of Daily Commercial Vehicle Trips by Hour
Since one of the objectives of understanding commercial vehicles is to identify the impact on peak periods, we have reviewed these data in typical a.m. and p.m. peak periods. These data are presented in Table 4.7. The a.m. peak period is 6:00 a.m. to 9:00 a.m. and the p.m. peak period is 3:00 p.m. to 6:00 p.m. The off-peak time period includes all remaining hours. The Atlanta, Denver, and Triad surveys show a consistent pattern where from 65 to 71 percent of commercial vehicle traffic occurs in the off-peak period, with 29 to 35 percent occurring in the peak period. The Atlanta and Denver surveys, representing larger urban areas, have 13 to 22 percent of traffic in individual peak periods, where the Triad survey has a wider range of commercial vehicles in the individual peak periods. This may be an anomaly of the Triad survey or may be indicative of patterns of commercial travel in smaller urban areas.
| Time Period | Atlanta | Denver | Detroit | Triad | Total |
|---|---|---|---|---|---|
| A.M. Peak | 13% | 15% | 48% | 3% | 31% |
| Off-peak | 65% | 71% | 50% | 69% | 58% |
| P.M. Peak | 22% | 14% | 3% | 28% | 11% |
The Freight Analysis Framework (FAF) data allow us to review intercity freight and non-freight (urban commercial) trucks by urban area. These data are presented in Table 4.8 as the sum of all urban areas as it is felt that the data are more robust in total than they would be for individual urban areas. These data show that freight trucks have a much higher percentage of VMT on freeways and lower percentage of total VMT on other facilities. Non-freight trucks have a similar percentage of VMT across all facility types, as expected, since these include more trips made to serve businesses and residences on local and arterial streets.
| Facility | Vehicle Miles Traveled: Auto | Vehicle Miles Traveled: Freight Truck | Vehicle Miles Traveled: Non-Freight Truck | Vehicle Miles Traveled: Total |
|---|---|---|---|---|
| Interstate | 213,326,538 | 5,085,094 | 12,024,381 | 230,436,013 |
| Expressway | 91,586,794 | 1,227,175 | 4,906,545 | 97,720,514 |
| Principal Arterial | 48,422,950 | 656,831 | 2,419,347 | 51,499,128 |
| Minor Arterial | 428,272 | 8,221 | 33,938 | 470,432 |
| Minor Collector | 57,575 | 204 | 4,354 | 62,132 |
| TOTAL | 353,822,128 | 6,977,526 | 19,388,565 | 380,188,219 |
| Facility | Percent of Total VMT: Auto | Percent of Total VMT: Freight Truck | Percent of Total VMT: Non-Freight Truck | Percent of Total VMT: Total |
| Interstate | 60.3% | 72.9% | 62.0% | 60.6% |
| Expressway | 25.9% | 17.6% | 25.3% | 25.7% |
| Principal Arterial | 13.7% | 9.4% | 12.5% | 13.5% |
| Minor Arterial | 0.1% | 0.1% | 0.2% | 0.1% |
| Minor Collector | 0.0% | 0.0% | 0.0% | 0.0% |
| TOTAL | 100.0% | 100.0% | 100.0% | 100.0% |
