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 6
2.1 Sources of Estimation Error To create a data-collection and analysis process that produces accurate axle-load distributions for pavement design, errors resulting from the following sources must be minimized: 1. The calibration of the data-collection equipment, 2. Differences in axle-weight distributions among different VCs, 3. Differences in vehicle loading rates between one road and another, 4. Differences in vehicle load by direction, 5. Variation in axle weights caused by changes in loading conditions by time of day, 6. Variation in axle weights caused by changes in loading conditions by day of the week, 7. Variation in axle weights caused by changes in loading conditions by time of year, and 8. Future changes in vehicle loading conditions. The first of these sources of error, calibration error, is a function of equipment operation. It is important that highway agencies calibrate their data-collection equipment carefully and use that equipment in locations where it can operate correctly. WIM equipment is particularly sen- sitive to calibration error, and calibration error is affected by a variety of factors, including sen- sor installation and condition, pavement roughness and condition, environmental conditions, and even roadway geometrics. Individual WIM sensor technologies are affected differently by these factors. Additional information about the selection, calibration, and operation of WIM equipment is contained in NCHRP Report 509.1 The second source of error, differences in weight among different VCs, is handled by sepa- rating the axle-weight data of one class of vehicles from those of other classes. Thus, separate axle-load distribution tables are required for each class of vehicles. VC may be defined by each highway agency. TrafLoad currently allows the use of the 13 standard FHWA VCs and/or an aggregation of these classes into a smaller set of user-defined classes. The latter option makes it possible to develop pavement designs for sites for which vehicle classification data are avail- able only by length class. To compute accurate load estimates, a highway agency must also collect vehicle classification volume data using VCs that can be correlated directly to the weight data that the agency col- lects. Chapter 3.0 provides instructions on how to collect and manipulate the data needed to determine the number of vehicles, by class, that will operate on the pavement being designed. The third and fourth sources of error arise because different roads can serve similar trucks that have very different sets of loading characteristics. This is also true for different direc- 1 Hallenbeck, Mark, and Herbert Weinblatt, NCHRP Report 509: Equipment for Collecting Traffic Load Data, Transportation Research Board of the National Academies, Washington, D.C., 2004. 2-6
OCR for page 7
tions of travel on the same road. These variations in loading characteristics are caused by dif- ferences in the commodities being carried and by differences in the percentages of trucks that are loaded. For example, on a road leading to a gravel pit, the lane of travel leading away from the pit will experience heavier trucks than the lane of travel leading to that pit, because most trucks leave the pit full but return to the pit empty. For this example, the same number of trucks and the same type of trucks operate in both directions, but the load experienced in each direction is very different. This same effect occurs when two parallel county arterials of simi- lar size carry very different heavy-vehicle loads. The loads are determined not by the size of the arterials but by the nature of the economic activity associated with each road. (For exam- ple, one arterial may lead through land zoned for residential use, while another leads through land zoned for warehouse use.) The next three sources of error, variation by time of day, day of week, and time of year, also occur because the kinds of trucks using a road can change during these time periods. Week- day and weekend traffic can be very different because of the presence or absence of truck traf- fic generated by businesses that are open only during weekdays. Similarly, vehicle weights can change by time of day. Lastly, a number of phenomena can cause changes in vehicle weights over the course of a year. For example, in agricultural areas trucks may be very heavy during some times of the year (e.g., harvest time) and lighter during others. In some states, legal weight limits change over the course of the year; loads can be legally increased during the winter months when pavement is strong but must be reduced during the spring-thaw months when the subgrade strength is low and heavy loads cause serious pavement damage. (Note that spring load restrictions are also a source of variation from one road to another, as some states only apply load restrictions to minor roads or to roads in specific geographic areas of the state.) Finally, loading rates can change over the years as weight laws change or as economic activ- ity affecting a roadway changes. The Pavement Design Guide software assumes that the shapes of the normalized axle-load distribution curves do not change from year to year, and the research team makes the same assumption in TrafLoad. The research team adopts this assumption both for consistency with the Pavement Design Guide software and because the research team cannot accurately forecast these types of changes. While it is possible to fore- cast changes in truck volumes on the basis of expected changes in economic activity, it is extremely difficult to forecast the effects such changes will have on axle-load distributions. Load distributions are a function of · Truck size and weight laws (will changes in regulations encourage heavier gross vehicle weights but cause those weights to be carried on more axles?), · The commodity characteristics of specific routes (do the commodities "cube out," or are loads limited by weight laws?), and · The fraction of loaded and unloaded trucks on the roadway (which is a function of both trucking fleet efficiency, types of haul, and the nature of truck ownership and use, all of which are difficult to forecast). This chapter deals exclusively with creating a data-collection program that, in conjunction with the TrafLoad software, helps an agency account for variations in vehicle loads by location, time 2-7