U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
REPORT |
This report is an archived publication and may contain dated technical, contact, and link information |
|
Publication Number: FHWA-HRT-13-091 Date: November 2014 |
Publication Number: FHWA-HRT-13-091 Date: November 2014 |
The vehicle classification algorithm for the refined LTPP WIM classification rule set was tested at the three SPS test site locations in Pennsylvania, Maryland, and Tennessee. To test the algorithm, researchers viewed digital photographs of passing trucks that had been classified with the newly installed rules and compared those images with the correct FHWA classification. The evaluation also determined how those trucks had been classified under the originally adopted LTPP WIM system algorithm. (Please note that the photographs in this chapter were taken by a remote, automatic camera in available light with a fixed field of view, which accounts for the blurring and altered color.)
The evaluation found that the WIM algorithm appears to work as intended. No errors were observed with the vehicles tested. The researchers conclude that the new algorithm should be installed at LTPP WIM sites. The results for each of the three test sites are described below.
The Pennsylvania WIM site produced relatively few photographs for review. Although vehicle records for multiple weeks of data were available, photographs were available for only 1 week: April 13—21, 2012. However, when used in combination with the data available from other sites, these data were sufficient for judging the performance of the revised LTPP WIM classification algorithm.
In lane 4 (the camera lane), 769 Class 7 trucks were observed from April 13 to April 21. Of those trucks, only three had six axles, and none had seven axles—the new Class 7 definitions being tested. All three of the six-axle Class 7 vehicles had been classified as Class 15 with the old algorithm. Figure 27 shows one of the three six-axle Class 7 trucks.
Figure 27. Photo. Six-axle Class 7 truck at Pennsylvania site.
Between April 13 and April 19, when the new algorithm was used in lane 4 at the Pennsylvania site, 288 Class 10 vehicles were observed. Of those, 32 vehicles had seven axles and 34 vehicles had eight axles—the two new categories of Class 10 trucks. All of the seven- and eight-axle vehicles had been classified as Class 13 vehicles under the old LTPP algorithm. They were correctly classified as Class 10 by the new algorithm. Unlike in many western States where Class 10 vehicles are often resource haulers (e.g., dump trucks) pulling pup trailers, all of the observed new Class 10 vehicles were conventional tractors pulling heavily loaded low-boy trailers, usually with an indivisible oversized load. Figure 28 and figure 29 provide good illustrations of these vehicles.
Figure 28. Photo. Image 1 of eight-axle Class 10 truck at Pennsylvania site.
Figure 29. Photo. Image 2 of eight-axle Class 10 truck at Pennsylvania site.
Twenty Class 13 vehicles were observed in lane 4 between April 13 and April 19. Of these vehicles, only three had 10 axles, one had 11 axles, and two had 12 axles (the new LTPP Class 13 definitions). None of these large vehicles had been classified by the old LTPP WIM algorithm. All of the very large vehicles were heavy-duty tractors pulling low-boy trailers with an additional articulated, king pin-equipped connection between the tractor and the load-carrying trailer. This additional connection typically allowed the use of an additional tridem set of axles to support the low-boy trailer, which was commonly carrying a heavy piece of machinery. Figure 30 shows one of these loads. (Note the two king pins, one over the tractor’s drive tandems and the other over the following tridems.) All were correctly classified as Class 13 vehicles.
At this Pennsylvania site, in the lane with the camera, 66 vehicles were changed from a vehicle class under the old LTPP algorithm that was valid (but incorrect) to a different—correct—vehicle class under the new LTPP algorithm. In addition, the total number of classified vehicles increased by nine in the LTPP test lane because these nine vehicles had been previously observed but not classified.
Figure 30. Photo. Eleven-axle Class 13 truck at Pennsylvania site.
Maryland data were tested for the 6 months, from July 2011 through December 2011. The site had fairly low volumes, so relatively few trucks of the new classes were observed even over this extended period.
Five seven-axle Class 7 trucks were observed in lane 1 (the camera-equipped lane) during this 6-month test. All were correctly classified as single-unit Class 7 trucks. All five trucks had been classified as Class 13 vehicles under the old LTPP WIM rule set. An example of the type of truck now correctly classified as a seven-axle Class 7 is shown in figure 31.
Figure 31. Photo. Seven-axle Class 7 truck at the Maryland site.
During this same period, 18 additional six-axle Class 7 trucks were observed in lane 1. All of these trucks had been classified as Class 15 vehicles. Therefore, they had not been included in the original LTPP load spectra. An example of a six-axle Class 7 truck in Maryland is shown in figure 32.
Figure 32. Photo. Six-axle Class 7 truck at the Maryland site.
These were a small percentage of the 475 Class 7 trucks of all axle configurations observed between July and December in lane 1.