Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 20


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 19
19 Most of the case study interviewees regarded traffic density comprehensive costs for traffic fatalities and injuries. Per as a major factor in crash risk. Carrier G uses truck routing capita congestion costs varied directly with city size. Per software which in its algorithms considers traffic characteris- capita crash costs varied inversely with city size. Among all tics in the vicinities of delivery locations. Carrier J, located in U.S. cities in the analysis, average per capita congestion cost upstate New York, monitors New York and surrounding state (in 2005 dollars) was $430. Per-person crash costs in those traffic alerts daily to warn drivers of congestion. same cities was $1,051. Thus, for the urban populations, crashes cause more than twice the economic loss (and asso- In a research partnership with the American Transporta- ciated harm) as does traffic congestion. tion Research Institute (ATRI), the FHWA Office of Freight Management and Operations has developed the Freight Per- formance Measures (FPM) program. FPM (www.freight EFFICIENT SCHEDULING: OPTIMAL TIMES FOR SAFE TRAVEL performance.org) provides extensive freight travel speed data for the U.S. highway system. Initial analyses have been Consider the ebbs and flows of vehicle traffic within the of speeds and travel time reliability on five major U.S. freight 24 hours of each day and the 7 days of each week. Almost all corridors: Interstates I-5, I-10, I-45, I-65, and I-70. Travel of us adapt our driving patterns to those variations in traffic speed data have been collected from more than 500,000 oper- density in an effort to travel quickly and efficiently. The speed ational trucks equipped with GPS-based automatic vehicle paradox described in chapter one suggests that when we seek location equipment. Trucks are assigned an anonymous iden- smooth, fast travel, we also find relatively safe travel. This in tification number to maintain the confidentiality of truckers turn suggests that evening and overnight driving would be and trucking companies. The system receives position (lati- safest because traffic is lightest at these times. A counter- tude and longitude) and time and date data from trucks at reg- argument is that night driving is inherently riskier because of ular intervals to provide data for the travel speed analysis. driving in darkness, the greater likelihood of driver fatigue, Trucks that stop (e.g., for refueling, deliveries, or rest) are and the greater presence of impaired motorists on the road- excluded from the calculations. ways. In the LTCCS, 62% of truck driver asleep-at-the-wheel crashes occurred in the 2-hour period between 4:01 a.m. and An FPM service ("FPMweb") allows carriers and other 6:00 a.m. This is well known to sleep researchers as a "circa- users to obtain information on travel speeds and delays for dian valley" (Knipling 2009). Alcohol use by other motorists any given place and time along 25 Interstate highways: I-5, is another major nighttime risk. One analysis found that more I-10, I-15, I-20, I-24, I-25, I-26, I-35, I-40, I-45, I-55, I-65, than one-third of fatal cartruck collisions during the overnight I-70, I-75, I-76, I-77, I-80, I-81, I-84, I-85, I-87, I-90, I-91, hours involved an alcohol-impaired car driver (Blower and I-94, and I-95. Users may generate Geographic Information Campbell 1998). System (GIS) maps, detailed analyses of individual corri- dors, or broader analyses across corridors. A strong majority of large-truck crashes and incidents occur during the daylight hours. Here are some percentages for day- One ATRI study (Short et al. 2009) used FPM data to iden- light (including dawn and dusk) crashes and traffic conflicts: tify the 30 worst freight bottleneck locations in the United States. This was based on FPM calculations of hourly and All 2008 police-reported crashes involving large trucks: total "Freight Congestion Value" for these locations. "Freight 79%; Congestion Value" was defined as the freight vehicle popula- 2008 fatal crashes involving large trucks: 68% (FMCSA tion times the average vehicle miles per hour below free flow Analysis Division 2010); (i.e., 55 mph). This was calculated hourly and in total. The LTCCS CT crash involvements: 73%; study did not include crash counts, but the evidence cited in LTCCS ST crash involvements: 90%; and this report suggests the same locations would be high-crash- CT naturalistic driving incidents (Hickman et al. 2005): risk as well. A more recent analysis lists 100 such sites in 75%. descending order (ATRI 2010). Across the 100 sites, the aver- age nonpeak-to-peak congestion ratio was 1.20. Unfortunately, crash and incident data alone do not provide satisfactory answers to the day-versus-night question. Crash The obvious benefit of avoiding congestion delays is databases precisely document crash times, but they have no the time savings. But is it the greatest benefit? An Ameri- corresponding exposure base to serve as a denominator for can Automobile Association study (Meyer 2008) does not generating relative crash rates by hour-of-day. squarely address the question posed, but does provide a per- spective on the overall costs of congestion versus those of Naturalistic driving studies do provide exposure data based crashes. The study compared the costs of crashes to the costs on onboard recordings of driving times and on randomly of congestion (for all vehicle types) by calculating a per- selected "exposure points." In the same CT naturalistic driving person cost for crashes and multiplying it by the population data cited earlier (Hickman et al. 2005), only 59% of driving figures in the same U.S. urban areas studied by TTI, as was during daylight, versus 75% of incidents. The odds ratio described previously. Crash costs were based on FHWA for incident occurrence during daylight versus darkness was