U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-RD-01-143
Date: October 2003
The CD-ROM contains three programs. The QC program is provided for each surface type-asphalt, jointed concrete, and continuously reinforced concrete. These programs will examine the time-series trend of all of the distress data for a given section.
The system requirements for running this software are:
Each setup program occupies a unique directory. The AC directory contains the setup for the DisQCAC program that examines the asphalt surface sections, the JC directory contains the setup for the DisQCJC program that examines the jointed concrete sections, and the CRC directory contains the setup for the DisQCCRC program that examines the continuously reinforced concrete sections.
The setup procedure is the same for all three programs. Select the directory corresponding to the program to be installed. Under that directory run SETUP.EXE. The setup program will prompt the user for the drive and directory to which the program is to be installed. Each of the three programs may be installed in the same directory. The program will automatically create icons under Programs in the Start menu.
For the program to run properly, it will be necessary to perform a data extraction of the distress data from the Regional Information Management System (RIMS). The extraction will need to consist of the MON_DIS_AC_REV, MON_DIS_PADIAS_AC, and MON_DIS_PADIAS42_AC (file extensions M10, M15, and M27 respectively) for the AC program. Extractions of tables MON_DIS_CRCP_REV, MON_DIS_PADIAS_CRC, and MON_DIS_PADIAS42_CRCP (file extensions M11, M16, and M28, respectively) are required for the CRC program. Extractions of tables MON_DIS_JPCC_REV, MON_DIS_PADIAS_JC, and MON_DIS_PADIAS42_JPCC (file extensions M08, M17, and M29, respectively) will be required for the JC program. The data extraction should use a fixed-width file format.
The output files are all named PASS_surface type.DAT and NOTPASS_surface type.DAT. For example, the outfiles are named PASS_AC.DAT and NOTPASS_AC.DAT for the asphalt-surfaced sections.
Start the program from the Start menu. The program will warn you that you need to select the location of the input data. This location is where the output data also is stored.
Several buttons and boxes will appear on the user interface. The top right button titled "···" is to be used for selecting the location of the input data files. Click on the "···" button and the interface will then allow you to select the drive and directory where the input data are stored. To select the desired drive and directory, it is necessary to click on a file contained within that directory.
At the bottom of the user interface is the "Process" button. This will initiate processing of the distress data. The first time the program is run, the software will have to combine the three data extraction files and then sort them by State code, SHRP ID, and survey date. This process will take a little time.
The process of sorting will not be required every time the program is run. The program will look for a file called S.DAT. If that file already exists, it will not reprocess the data files. If the user wishes to run the program from the beginning with a new data extraction file, the user needs to delete the following files from the data input directory: D.DAT, S.DAT and S2.DAT (if S2.DAT exists).
After the program has sorted the data, it begins checking each of the selected distresses with a unique State code, SHRP ID, and construction number. The first step of the software check process is a linear regression. It then determines if the predicted values from the regression fall within three COVs of the actual data point (explained in chapter 2). The distresses and COVs used are shown in table 5. The COV is multiplied by the measured distress. This value is the standard deviation used for the quantity of the distress. The standard deviation is multiplied by three, then added to and subtracted from the mean. If the regression line falls within that range, then that data point passes. If all of the data points pass this check, the distress for that unique State code, SHRP ID, and construction number are written to a file called PASS*.DAT, where * is the surface type.
If one survey date fails this check, then a series of other checks are completed. The first of these examines the maximum recorded value of that distress. If that value is less than 5 lineal meters of distress, less than 2 m2 of area distress, or less than 2 for a distress that is counted such as number of cracks, then that distress is considered minimal. In this case, the distress is placed in the PASS*.DAT file.
Next, the software checks to determine if the amount of distress increases with each consecutive survey date. It is anticipated that the distress will increase and fall along a straight line. If it is increasing, but not linearly, the distress is said to be increasing exponentially. This distress for a unique State code, SHRP ID, and construction number are written to the PASS*.DAT file.
If only the first or the last points fall outside the limits, then the software checks determine which side of the line these points fall on. If the first data point falls below the regression line or the last data point falls above the regression line, then the distress data for that unique State code, SHRP ID, and construction number are written to the PASS*.DAT file. Any data set that fails all of these checks is considered discrepant and will be written to the NOTPASS*.DAT file. The program will attempt to determine why the discrepancy has occurred.
In the review of the distress, the most common identifiable discrepancies were differences in the classification of the distress between surveys. The program checks for: a) offsetting discrepancies between fatigue and longitudinal cracking in the wheel path, b) offsetting discrepancies between block cracking and transverse and longitudinal cracking, and c) offsetting discrepancies between longitudinal cracking in the wheel path and non-wheel path longitudinal cracking. In these cases, an offsetting discrepancy indicates that the actual data point lies above the regression line for one distress, and for the same survey the observation for the other distress lies below the regression line. These surveys will have comments identifying these potential causes in the NOTPASS*.DAT file.
If the cause of the discrepancy is unidentifiable, then the message "Error Undeterminable" is written to the NOTPASS*.DAT file along with the State code, SHRP ID, and construction number. For that section and distress, none of the data will be included in the PASS*.DAT file.
Two other checks are performed on data. These checks concern only transverse cracking and patching. The average length of transverse cracks is determined by dividing the total length of transverse cracking by the number of transverse cracks. If the average crack length is less than 0.3 meters (m) or longer than 3.7 m, then the survey is recorded in the NOTPASS*.DAT file.
The patching area is divided by the number of patches. The average patch size is checked to ensure that it is larger than 0.1 m2 and smaller than 90 m2. Just as with the transverse cracking, that distress for that survey is written to the NOTPASS*.DAT file along with an "Apparent Summarization Error" message. That distress for that survey is kept out of the PASS*.DAT file.
The only time a single survey will be recorded as discrepant is when that survey represents less than 20 percent of the data for a given section. In this case, none of the data for a particular distress on the section of interest will be written to the PASS*.DAT file. However, the single discrepant survey will be included in the NOTPASS*.DAT file. When this bad data set is verified, it can be excluded from future reviews (as noted below) and the good data sets will then be added to PASS*.DAT.
If only one data point is available for a given construction event on a section, then that survey is automatically passed to the PASS*.DAT file.
In the event that only two surveys are present, then the slope of the line connecting those two points is checked. If the slope is positive, then the data are written to the PASS*.DAT file. If the slope is negative, then the data are still written to the PASS*.DAT file and also to the NOTPASS*.DAT file with a message stating that only two points existed and that it had a negative slope.
When the data are completely processed, the user will need to review and revise any errors that have occurred, if possible. Some of these discrepancies may be irreparable. A distress survey may be identified as errant, but irreparable. In this instance, the user may wish to keep a particular survey from being included in future reviews.
The boxes on the left side of the user interface allow the user to enter a section, survey date, and distress to exclude from the review. To complete the process, the user must click on the exclude button at the bottom left side of the window. The print button will allow the user to print a list of section IDs, survey dates, and distresses that were previously excluded.
If the user is able to reconcile some of the discrepancies that were previously considered irreparable, it is possible to delete one of the distresses that was excluded. The section ID, survey date, and distress should be entered in the boxes on the right side of the interface. To complete the process, press the delete button on the bottom right side of the interface. This will delete a distress for a given section from the list of distresses that are being excluded from review.