Traffic Data Collection, Analysis, and Forecasting for Mechanistic Pavement Design (2004) / Chapter Skim
Currently Skimming:
Pages 60-116
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 60... ...
Part 2 Guidelines for Collecting Traffic Data to Be Used in Pavement Design
|
From page 61... ...
poor information about the truck volumes and axle loads to be applied, with Level 3 further divided into Levels 3A and 3B. Corresponding to each input level, the Pavement Design Guide software requires site-specific, region-specific, or default values for several types of traffic data.
|
From page 62... ...
and/or an aggregation of these classes into a smaller set of user-defined classes.3 The latter option makes it possible to develop pavement designs for sites for which vehicle classification data are available only by length class. 1.2 The Remainder of Part 2 Chapter 2.0 of Part 2 presents in some detail the development of a program for collecting the weight data that TrafLoad requires for generating axle-load distribution factors and estimates 2-2 3 The Pavement Design Guide software currently allows the use of up to 13 classes.
|
From page 63... ...
The final chapter discusses the collection and handling of the traffic data required to create the necessary datasets. 2-3 Table 1.2 FHWA Vehicle Classes 1.
|
From page 64... ...
For any pavement project, the required load spectra are developed by TrafLoad using data collected from WIM equipment either on the same road at a site reasonably near the pavement project (Pavement Design Guide Level 1 data) , or from WIM data collected elsewhere (Levels 2 and 3)
|
From page 65... ...
2-5 Table 2.1 Load Ranges Used for Load Spectra Upper Limit of Load Range (kipsa) by Type of Axle Group Load Range Single Tandem Tridem Quad 1 3 6 12 12 2 4 8 15 15 3 5 10 18 18 12 21 21 14 24 24 16 27 27 18 30 30 20 33 33 22 36 36 24 39 39 26 42 42 28 45 45 30 48 48 32 51 51 34 54 54 36 57 57 38 60 60 40 63 63 42 66 66 44 69 69 46 72 72 48 75 75 50 78 78 52 81 81 54 84 84 56 87 87 58 90 90 60 93 93 62 96 96 64 99 99 4 6 5 7 6 8 7 9 8 10 9 11 10 12 11 13 12 14 13 15 14 16 15 17 16 18 17 19 18 20 19 21 20 22 21 23 22 24 23 25 24 26 25 27 26 28 27 29 28 30 29 31 30 32 31 33 66 102 102 32 34 68 33 35 70 34 36 72 35 37 74 36 38 76 37 39 78 38 40 80 39 41 82 a One kip = 1,000 pounds = 4.448 kN.
|
From page 66... ...
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 direc1 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.
|
From page 67... ...
While it is possible to forecast 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?
|
From page 68... ...
is addressed in the next chapter, and equipment calibration is covered in a companion report.2 Seasonal variation in axle weights are of particular interest in pavement design because the pavement damage caused by axle loads is significantly affected by seasonal variations in soil conditions such as wetness, freezing, and thawing. As a result, the Pavement Design Guide software is designed to use separate sets of load spectra for each month of the year and to incorporate the effects of seasonal variations in the load spectra in the resulting pavement designs, and TrafLoad has been designed to produce such seasonally varying load spectra.
|
From page 69... ...
The software uses seasonal load spectra datasets that contain data for the 12 months of the year as the basis for imputing seasonal adjustments to data collected for periods of less than 12 months. The process for performing these adjustments (presented in Part 4, which is available online at http://trb.org/news/blurb_detail.asp?
|
From page 70... ...
Recent analyses of California data indicate that, for combination trucks, TWRGs produce mean absolute percentage errors (MAPEs) for pavement stresses (as measured in 18,000pound equivalent single-axle loads)
|
From page 71... ...
2-11 Figure 2.1 Load Distributions for Tandem Axles of FHWA Class 9 Trucks at Three Different Sites Figure 2.1(a) Lightly Loaded Trucks Fraction of Tandem Axles in Weight Group 0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 Figure 2.1(b)
|
From page 72... ...
(This distribution was obtained by averaging the three load distributions shown in Figure 2.1.) Statewide distributions are the least reliable load spectra for pavement design.
|
From page 73... ...
First, for any month for which all seven DOW sets of load spectra are available, the monthly load spectra are developed by averaging the seven sets of load spectra in a way that avoids overweighting or underweighting the load spectra collected on any particular day of the week.5 For all months for which all seven DOW load spectra are available, this form of averaging Figure 2.2 A Typical Statewide Load Distribution for FHWA Class 9 Trucks 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 Fraction of Axles in Weight Range Axle Weight Groupa a Each group is identified by the maximum weight, in thousands of pounds, for the group. 5 For each VC, the averaging procedure uses the monthly average DOW traffic volumes for that class to obtain weighted averages of the seven DOW load spectra for the class.
|
From page 74... ...
Seasonal load spectra datasets are developed using data collected from a single lane (or design lane) using WIM equipment that has been consistently calibrated over the entire collection period.
|
From page 75... ...
For each of these groups, the software will use the seasonal load spectra datasets to develop a set of monthly adjustment factors. The assignment of WIM sites to seasonal load spectra factor groups is based on how average pavement damage per vehicle varies over the course of a year.
|
From page 76... ...
TrafLoad also allows the user to associate any site for which a seasonal load spectra dataset does not exist with a site for which such a dataset does exist and at which seasonal and DOW variations in load are believed to be very similar to those for the site in question. If such an association is made, daily and monthly adjustment factors developed using data from the associated site will be applied to data from the site in question in the absence of factors developed from a seasonal load spectra dataset for that site.
|
From page 77... ...
Pavement designs for such a site are developed using either a set of default load spectra developed on a statewide basis or a default set developed for a particular TWRG to which the site has been assigned. The goal of assigning sites to TWRGs is to permit the use of load spectra that better describe the axle loads at member sites than would a statewide set of load spectra.
|
From page 78... ...
should be assigned to different TWRGs than roads on which axle-weight limits do not vary seasonally. A corollary to the above principle is that the WIM sites whose data are used for deriving load spectra for a given TWRG need not all be in the same state, provided that they are all subject to the same weight limits and that the trucks operating at the out-of-state WIM site(s)
|
From page 79... ...
On the other hand, as the number of WIM sites in a TWRG grows, opportunities also grow for splitting the TWRG to produce smaller TWRGs, each with more uniform sets of load spectra. 7 Additional discussion of the formation of TWRGs is contained in FHWA's Traffic Monitoring Guide (May 2001, Section 5, Chapter 3)
|
From page 80... ...
If the only Level 1 WIM sites assigned to this TWRG are the two sites mentioned above, the load spectra developed for this TWRG will be formed as unweighted averages of those developed for the two sites. (The current version of TrafLoad does not allow for using weighted averages in developing load spectra for TWRGs.)
|
From page 81... ...
Unfortunately, both the cost of WIM data-collection and functional limitations in WIM sensor technology restrict the number and location of data-collection points at which state highway agencies can collect WIM data. WIM equipment only works accurately on flat, smooth pavements that are in good condition.9 In addition, each time a WIM scale is placed in or on a pavement, the effects of road profile and roughness on vehicle dynamics mean that the scale must be recalibrated in order to collect data accurate enough to be used as input to the pavement design process.
|
From page 82... ...
(Note that these sites also provide continuous classification data as well as continuous volume data and thus take the place of ATRs and permanent vehicle classifiers.) Analyses of these continuous data sources allow states to learn if truck weights are changing over time or if they change by season of year or even by time of day.
|
From page 83... ...
The second section provides a brief summary of the traffic data produced by TrafLoad for use by the Pavement Design Guide software. Sections 3.3–3.5 discuss in some detail the classification data that are required by TrafLoad in order to produce these outputs.
|
From page 84... ...
Thus, it may be practical to collect Level 1 classification data at some sites for which only Level 2 weight data are available and vice versa. 3.2 Data Produced for the Pavement Design Guide Software TrafLoad produces a moderate amount of traffic data for Level 1, 2, and 3A classification sites and a very limited amount of data for Level 3B classification sites.
|
From page 85... ...
Continuous Classification Counts The principal goal of the continuous classification-count program is the creation of factors needed to estimate annual average daily truck volumes from short-duration classification 2-25 1 For sites for which HDFs are not provided to the Pavement Design Guide software, the software uses its own set of HDF default values.
|
From page 86... ...
To accomplish this goal, it is necessary to measure day-of-week and seasonal variation in truck traffic and to develop factors that can be applied to short-duration counts. As illustrated by Figures 3.1 and 3.2, truck volumes vary significantly by time of day (TOD)
|
From page 87... ...
The trend in truck volume should be examined both for each individual roadway on which a permanent data-collection device is located and for each of the geographic areas in the state. (Truck traffic tends to vary with both the economic activity taking place in a geographic region and the amount of through traffic passing through the area.)
|
From page 88... ...
In particular, current-year traffic ratios will provide a better adjustment for any unusual conditions affecting truck volumes at the time that a short-duration count is collected. (Such conditions include unusual weather, a poor harvest, or the beginning of a sharp recession or of an economic recovery.)
|
From page 89... ...
DOW factoring is designed to correct for the overestimates that are reflected in weekday classification counts. In order to use DOW factoring effectively, it is necessary to distinguish VCs and sites where business-week trucking predominates from VCs and sites where through trucking predominates.
|
From page 90... ...
In TrafLoad, VC groups used in the factoring of classification counts are called "Type 1 VC groups" to distinguish them from the WIM VC groups discussed in the preceding chapter. The Type 1 VC groups are also used in the development and application of seasonal and TOD traffic ratios.
|
From page 91... ...
If this is not done, there is a small possibility that the TrafLoad factoring procedure will be forced to terminate abnormally in order to avoid dividing by zero.3 Seasonal Factor Groups As discussed above, the DOW factor groups should be designed to group sites with similar DOW patterns of truck volume. Similarly, the seasonal factor groups should be designed to group sites with similar seasonal (or month-of-year)
|
From page 92... ...
for each user-defined TOD factor group. Since partial-day classification counts are almost always collected on a weekday, only weekday classification counts are used in developing TOD traffic ratios.
|
From page 93... ...
Hence, TrafLoad uses combined traffic ratios for factoring counts from Level 1B sites and separate monthly and DOW traffic ratios, developed using data from groups of Level 1A sites, for factoring counts from Level 2 sites.
|
From page 94... ...
at a relatively fixed set of sites to provide general information about truck volumes and how these volumes are changing over time; 2. Expected project counts, collected at sites at which highway projects are anticipated, to provide data for use in the planning and design process; and 3.
|
From page 95... ...
Level 2A classification sites are sites at which automatic vehicle classifiers (AVCs) are used to obtain one or more classification counts over the course of a year.
|
From page 96... ...
The quality of the estimates of AADTi that are produced will tend to vary with the degree to which the seasonal and DOW patterns in truck volumes at the Level 2A site match the volume patterns in the seasonal and DOW factor groups to which the site has been assigned. Use of 7-day counts reduces or eliminates the need for DOW factoring, and use of multiple counts over the course of a year reduces the need for seasonal factoring.
|
From page 97... ...
For such sites, TrafLoad uses the estimates of AADTi by direction at the associated site to distribute the user-supplied estimate of AADTT over the various truck 2-37 7 There is a relationship between the hourly distribution factors (HDFs) required by the Pavement Design Guide software and the TOD traffic ratios (TODTRs)
|
From page 98... ...
2-38 Table 3.2 Hourly Distribution Factors Used by TrafLoad for Level 2B Sites with Business-Day Trucking Hour Hourly Distribution Factor 0 0.6% 1 0.4% 2 0.4% 3 0.4% 4 0.9% 5 2.8% 6 4.8% 7 6.1% 8 7.4% 9 7.8% 10 7.7% 11 7.6% 12 7.5% 13 7.9% 14 8.0% 15 7.0% 16 5.9% 17 4.7% 18 3.6% 19 2.6% 20 1.9% 21 1.6% 22 1.3% 23 1.1% Derived from data for urban other principal arterials (Functional System 14) in Mark Hallenbeck, et al., Vehicle Volume Distributions by Classification, Chaparral Systems Corporation and Washington State Transportation Center, June 1997, for FHWA, FHWA-PL-97-025, pp.
|
From page 99... ...
The Pavement Design Guide software uses the TTCs as the basis for disaggregating estimated AADTT into the standard FHWA VCs. Pavement designs developed by the Pavement Design Guide software for Level 3B classification sites require the use of load spectra for the standard FHWA VCs.
|
From page 100... ...
A Simple Procedure A simple procedure for forecasting the rates of change in traffic volumes for the design lane or design direction at any particular project site is presented below. In the procedure, the rates of change are referred to as "growth rates" to emphasize that, for the purpose of pavement design, traffic growth is of primary interest.
|
From page 101... ...
4. Use regression to estimate either linear growth rates or exponential growth rates for each Level 1A site for each VC group.12 In choosing between the two types of growth, a simple option is to choose the type that is believed to best describe expected future growth in truck traffic at the project site -- linear growth if it is believed that the annual increase in this traffic is not likely to grow and exponential growth if this annual increase is expected to grow.
|
From page 102... ...
. The upgraded facility has attracted a significant amount of truck traffic heading northwest from Albuquerque that formerly used several other Interstate and U.S.
|
From page 103... ...
. A third forecast assumes the same truck volumes in the base year and in the forecast year as those of the second forecast, but the third forecast assumes linear growth.
|
From page 104... ...
This forecast is applied to total two-way truck volume, with no distinctions by lane, direction, or VC. 2-44 Table 3.4 TrafLoad Input Options for Forecasts User Inputs Input Option Linear Growth Exponential Growth Annual Annual change Annual percentage change Overall change Total change over period*
|
From page 105... ...
In addition, some state highway agencies may have to create new summary 2-45 1 Cambridge Systematics, Inc., and Washington State Transportation Center, Equipment for Collecting Traffic Load Data, prepared under NCHRP Project 1-39, June 2003 available online at http://trb.org/ news/blurb_detail.asp?
|
From page 106... ...
However, the availability and quality of data collected by each state will have a direct impact on the accuracy of traffic load inputs to the pavement design process and consequently on the reliability of the pavement designs developed with the new software. The basic data-collection design for providing traffic load data fits within the general traffic data-collection guidance provided by the FHWA in the 2001 Traffic Monitoring Guide.
|
From page 107... ...
, • Summarization of the data into statistics and record formats that can be readily used by others inside and outside the state highway department, and 2-47 Table 4.1 Data Required by the Pavement Design Guide Software Required Data Source for Data AADTi a for up to 13 VCs (1, 2, and 3A) b Continuous classification counts, or Short-duration classification counts adjusted for day of week and season AADT and Percent Trucks (3B)
|
From page 108... ...
sites Monthly traffic distribution factors by VC Trend measurements used when forecasting future truck volumes Short-Duration WIM Measurements Current load spectra datasets (Weight Level 1) (if a wellcalibrated site)
|
From page 109... ...
5 Cambridge Systematics, Inc., and Washington State Transportation Center, Equipment for Collecting Traffic Load Data, prepared under NCHRP Project 1-39, June 2003, available online at http://trb.org/ news/blurb_detail.asp?
|
From page 110... ...
Data Summarization Once the collected traffic data have successfully passed through the quality assurance process, an efficient mechanism is needed for storing and summarizing the data so that they can be used when needed for pavement design. Most states have existing programs that collect and store both volume and classification data on a section-by-section or count-by-count basis.
|
From page 111... ...
patterns of truck volumes; • TOD distributions for truck volumes; • Load spectra for different roads and roadway groups; and • Numbers of axles, by type of axle, for each class of trucks. The last two statistics have been discussed in some detail in Chapter 2.0, and other needed statistics (including the first two)
|
From page 112... ...
This separation limits interaction between these two groups and reduces the ability of the traffic data-collection group to adopt procedures that satisfy the changing input requirements of pavement design and that meet other needs of pavement designers. Most state highway agencies already collect the data needed for estimating traffic loads for mechanistic design.
|
From page 113... ...
Although state highway agencies are already doing much of what is needed to meet the traffic data requirements of mechanistic pavement design, considerable work is still required to refine the existing procedures and software. Although data are collected, they often are not adequately summarized and reported.
|
From page 114... ...
AADTT Annual average daily truck traffic (all classes, combined)
|
From page 115... ...
2A Site for which an AVC count is available for a period of at least 48 hours. 2B Site for which a manual classification count for a minimum of 6 weekday hours is available.
|
From page 116... ...
3 All other WIM sites.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.