Distress Data Consolidation (RD-01-143)
Final Report
508 Compliance Text
October 27, 2003

Figures

Figure 1. Illustration of user interface for graphical review software. Screen capture.
This figure is an example of the software developed for this study and illustrates the menu and box selections. The figure is a screen capture showing "Transverse Crack Number." On the upper left side are three boxes: the "State" box, displaying "38," the "SHRP ID" box, displaying "2001," and the "Distress Type" box, reading "Transverse Number." The "Construction Dates" box is on the right, parallel to the "Transverse Number" box, and shows the date of August 17, 1990. Above this box on the upper right are eight boxes in two columns, with the following headings: "Fatigue," "Long Crack WP," "Long Crack NWP," Trans Crack No.," "Trans Crack Len," "Block," "Patch Area," and "Patch No." All these boxes are blank. Below all of the above is a large box that contains a graph with four test methods, which are "Manual," "PADIAS (PADIAS) 4.2," PADIUS 1.X," and "High-Low." The date is graphed on the horizontal axis from June 1988 to June 1995. Transverse Crack Number is graphed on the vertical axis from 0 to 50. In June 1988, High-Low tested 18 and 38 cracks, and Manual tested 28 cracks. In 1989, PADIAS 1.X tested 28 cracks. In 1994, High-Low had 19, 20, 36, and 40 transverse cracks. PADIAS 4.2 had 25 cracks, and Manual had 30 cracks. In June 1995, High-Low had 20 and 45 cracks, and PADIAS 4.2 had 32 cracks.

Figure 2. Normal Probability Plot. Graph.
The figure is a bell curve with a mean of zero. The plot points range from negative 4 to positive 4. The curve extends 3 standard deviations from the mean. The areas beyond positive and negative 3 are both .5 percent of the total area.

Figure 3. Illustration of section that has been through the review. Screen capture.
The figure is a large screen capture of "Long Crack Non-Wheel Path (NWP)." The screen is formatted similar to the screen in figure 1. On the upper left side, the "State" box displays 01. The "SHRP ID" box displays 0101. "Distress Type" is "Long Crack NWP." In the upper right of the screen, the "Fatigue" box shows "Good," the Long Crack WP shows "Summary," and the "Long Crack NWP" box shows "Method." The discrepancy in the longitudinal wheel path cracking is a summarization error, and the discrepancy in the longitudinal non-wheel path cracking is due to data collection methodology. The graph is a scatter plot with three methods. The methods are "Manual," "PADIAS 4.2," and High-Low. The date is plotted on the horizontal axis from August 1994 to April 1998. Long Crack NWP, from 0 to 200, is graphed on the vertical axis. The Manual method is graphed at the lowest long crack NWP (0) throughout the year. High-Low site is plotted on March 1996 with 40 long cracks and 175 long crack NWPs. PADIAS 4.2 is on 100 long crack NWPs in March 1996.

Figure 4. Example of output file from graphical review software. Data table.
This figure shows how data collected is logged. The data file has six columns; Methods, State SHRP, Construction Event Number (CN), Survey (date), Value (or amount of distress), and Distress (type). The file has 11 lines under each of the columns. Line 1: ac underscore rev; 01; 0102; 1; 94-08-25; 0; fatigue. Line 2: ac underscore rev; 01; 0102; 1; 95-02-08; 1.8; fatigue. Line 3: ac underscore rev; 01; 0102; 1; 95-07-27; 0; fatigue. Line 4: pad42 underscore ac; 01; 0102; 1; 96-01-10; 0; fatigue. Line 5: ac underscore rev; 01; 0102; 1; 96-04-17; 2.1; fatigue. Line 6: ac underscore rev; 01; 0102; 1; 96-10-11; 26.3; fatigue. Line 7: ac underscore rev; 01; 0102; 1; 97-10-30; 21.1; fatigue. Line 8: ac underscore rev; 01; 0105; 1; 94-08-25; 0; long crack WP. Line 9: ac underscore rev, 01, 0105, 1, 95-02-08; 0; long crack WP. Line 10: pad42 underscore ac; 01; 0105; 1; 96-01-10; 0; long crack WP. Line 11: ac underscore rev; 01; 0105; 1; 96-04-19; 0; long crack WP. Two notes under the data read, "ac underscore rev equals manual survey," and pad42 underscore ac equals PADIAS 4.2 survey."

Figure 5 grapha A and B. Offsetting discrepancies for fatigue and longitudinal cracking noted on survey of January 1996. Graphs.
The figure is a comparison of two graphs that illustrate various discrepancies between fatigue and longitudinal cracking. Graph A. Fatigue discrepancy noted in survey of January 1996. The date is graphed on the horizontal axis from January 1990 to May 1997. Fatigue cracks are graphed on the vertical axis from 0 to 300. The figure has four methods, which are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. Manual and High-Low have significant fatigue in 1994 and 1997, with 75 and 180 fatigue cracks in 1994, and 150 and 280 fatigue cracks in 1997, respectively. In January 1996, PADIAS 1.X had fatigue cracks of 0 and 40, and PADIAS 4.2 had 20.

Figure 6. Illustration of offsetting discrepancies in longitudinal cracking caused by lack of distinction between wheel path and non-wheel path cracking. Graphs. Two graphs are compared.
Graph A. Discrepancy in longitudinal cracking in the wheel path noted in May 1991 survey. The date is graphed on the horizontal axis from April 1990 to September 1993. Longitudinal Crack WP is graphed on the vertical axis from 0 to 400. The three methods are Manual, PADIAS 1.X, and High-Low. High-Low has the highest Long Crack WP on April 1990, with 10 and 300 longitudinal cracks in the wheel path. Manual has 150 and PADIAS 1.X had 0 longitudinal cracks in mid-1991. Graph B. Discrepancy in non-wheel path longitudinal cracking noted in May 1991 survey. The date is graphed on the horizontal axis, from April 1990 to September 1993. Longitudinal Crack NWP is graphed on the vertical axis from 0 to 600. The methods are Manual, PADIAS 1.X, and High-Low. In July 1991, High-Low has 120 and 540 longitudinal cracks in the NWP. Manual has 0, and PADIAS 1.X has 340 longitudinal cracks. There were more longitudinal cracks in the NWP.

Figure 7. Illustration of discrepancy caused by lack of distinction between block cracking and transverse and longitudinal cracking. Graphs.
This figure shows discrepancies caused by lack of distinction between block cracking, transverse cracking, and longitudinal cracking. It contains five graphs.
Graph A. Discrepancies noted in block cracking in surveys of March 1993 and April 1996. The date is graphed on the horizontal axis from July 1988 to April 1996. Block cracking is graphed on the vertical axis from 0 to 1000. The discrepancies are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. There are discrepancies noted significantly on March 1993 and April 1996.In March 1993, High-Low has 300 and 900 block cracks, and PADIAS 4.2 tested 600 block cracks. In April 1996, High-Low tested 300 and 850 block cracks, and PADIAS 4.2 tested 580 block cracks. Graph B. No discrepancies noted on longitudinal cracking in the wheel path. The date is graphed on the horizontal axis from July 1988 to April 1996. The discrepancies are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. Longitudinal Crack WP is graphed on the vertical axis. All methods found no longitudinal cracks on the wheel paths. There are no discrepancies. Graph C. Discrepancies noted in non-wheel path longitudinal cracking noted on August 1989, March 1993, and April 1996. The date is graphed on the horizontal axis from July 1988 to April 1996. Long Crack NWP is graphed on the vertical axis from 0 to 250. The discrepancies are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. The discrepancies are significant on August 1998, March 1993 and April 1996. August has 50 and 250 longitudinal cracks for High-Low, and 150 for PADIAS 1.X. PADIAS 4.2 tested zero longitudinal cracks for March 1993 and April 1996. Manual, PADIAS 1.X, and High-Low had greater discrepancies than PADIAS 4.2. Graph D. Discrepancies noted in number of transverse cracks noted on August 1989, March 1993, and April 1996. The date is graphed on the horizontal axis from July 1988 to April 1996. The Transverse Crack Number is graphed on the vertical axis from 0 to 100. The discrepancies are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. In August 1989, High-Low tested 45 and 95 transverse cracks, and PADIAS 1.X tested 70. PADIAS 4.2 tested zero transverse cracks on March 1993 and April 1996. There are similar discrepancies found in transverse and longitudinal cracks for NWPs. Graph E. Discrepancies noted in length of transverse cracking. The date is graphed on the horizontal axis from July 1988 to April 1996. Transverse Crack Length is graphed on the vertical axis from 0 to 150. The discrepancies are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. High-Low and PADIAS 1.X tested greater transverse cracks in 1989 and 1990; in both years, High-Low tested 70 and 125 transverse cracks, and PADIAS 1.X tested 100 cracks. PADIAS 4.2 had zero transverse cracks in 1993 and 1996.

Figure 8. Illustration of discrepancy in length of transverse cracks caused by a summarization error. Graphs.
This figure contains two graphs compared with four sites. The methods for both graphs are Manual, PADIAS 4.1, PADIAS 1.X, and High-Low. Graph A. No Discrepancy noted in number of transverse cracks. The date is graphed on the horizontal axis, from June 1988 to June 1995. Transverse Crack Number is graphed on the vertical axis from 0 to 50. All methods have high transverse cracks throughout the graph. High-Low tested the highest and lowest count; highs around 42 and lows around 15. PADIAS 1.X, 4.2, and Manual ranged between the High-Low method, approximately 25 to 32 cracks. Graph B. Discrepancy in length of transverse cracks in survey of June 1988. The date is graphed on the horizontal axis from June 1988 to June 1995. Transverse Crack Length is graphed on the vertical axis from 0 to 200. Manual tested zero crack length in June 1988. The other methods had high lengths throughout the years. A summation error caused the discrepancies seen in figure 8.

Figure 9. Illustration of discrepancy in area of patching caused by summarization error. Graphs.
Two graphs are compared. The discrepancies in both graphs are Manual, PADIAS 1.X, and High-Low. Graph A. Discrepancies in number of patches noted in surveys of September 1993 and May 1996. The date is graphed on the horizontal axis from September 1989 to May 1996. Patch Number is graphed on the vertical axis from 0 to 25. On September 1993, Manual tested zero patches. On May 1996, High-Low tested 3 and 5 patches, and Manual tested 4 patches. There is a discrepancy between the two days. Graph B. Patch area in survey of May 1996 greater than 15 percent of total surface area of section. The date is graphed on the horizontal axis from September 1989 to May 1996. Patch area is graphed on the vertical axis from 0 to 800. On May 1996, High-Low tested 250 and 700 patch areas. For the remaining years, the three methods tested zero patch areas, which causes the discrepancy. The patch area for this date makes up more than 15 percent of total surface area. Both graphs have a discrepancy that may be caused by a summation error.

Figure 10. Illustration of discrepancy caused by lack of distinction between wheel path and non-wheel path longitudinal cracking in PADIAS 1.X data. Graphs.
Two graphs are compared. The discrepancies in both graphs are Manual, PADIAS 1.X, and High-Low. This figure illustrates discrepancies from lack of distinction between wheel and non-wheel paths. Graph A. Discrepancy noted in longitudinal cracking in the wheel path in the PADIAS 1.X survey of November 1991. The date is graphed on the horizontal axis from January 1990 to June 1995. Longitudinal Crack WP is graphed on the vertical axis from 0 to 150. PADIAS 1.X tested zero longitudinal cracks on the wheel path on November 1991. Graph B. Discrepancies noted in non-wheel path longitudinal cracking for all PADIAS 1.X surveys. The date is graphed on the horizontal axis from January 1990 to June 1995. Longitudinal Crack NWP is graphed on the vertical axis from 0 to 25. PADIAS 1.X tested 14 longitudinal cracks on January 1990, 15 cracks on December 1990, and 13 cracks on November 1992.

Figure 11. Illustration of discrepancy caused by insufficient quantities of distress on an asphalt-surfaced section. Graph. The date is graphed on the horizontal axis from August 1988 to March 1993. Transverse Crack Length is graphed on the vertical axis from 0.0 to 1.0. There are four methods being tested. The methods are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. In August 1992, High-Low tested 0.5 and 0.9 crack length, and PADIAS 4.2 tested 0.7 crack length. The other methods tested zero. There is a discrepancy caused by insufficient quantities of distress.

Figure 12. Illustration of discrepancy caused by insufficient quantities of distress on a jointed concrete-surfaced section. Graph.
This figure illustrates discrepancies from insufficient numbers of distress on jointed concrete surfaces. The date is graphed on the horizontal axis from November 1988 to March 1996. Longitudinal Crack is graphed on the vertical axis from 0.0 to 3.0. The methods are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. In 1992, High-Low tested 1.0 and 2.7 longitudinal cracks, and PADIAS 4.2 tested 1.5. The other methods tested zero throughout the years.

Figure 13. Illustration of discrepancy caused by insufficient quantities of distress on a continuously reinforced concrete-surfaced section. Graph.
This figure illustrates discrepancies caused by lack of distress on continuously reinforced concrete surface. The Survey Date is graphed on the horizontal axis from August 11, 1987 to July 24, 1998. Distress is graphed on the vertical axis from 0 to 0.25 meters. The methods are Manual, PADIUS, PADIUS 4.2, LL, and UL. PADIAS UL tested zero on 1988 and 1990. Manual tested zero on 1991, 1993, 1995, and 1997. On August 1995, PADIAS LL tested 0.175 and PADIAS UL tested 0.217.

Figure 14. Illustration of a discrepancy caused by an exponential growth rate. Graph.
The date is graphed on the horizontal axis from June 1990 to September 1997. Fatigue is graphed on the vertical axis from 0 to 500. The methods are Manual, PADIAS 4.2, PADIAS 1.X, and High-Low. In 1993, High-Low tested 0 and 50, and Manual tested 25. In January 1996, High-Low tested 0 and 50, and PADIAS 4.2 tested 25. In September 1997, High-Low tested zero and 450, and Manual tested 250. There was a vast increase in fatigue cracks in 1997.

Figure 15. Illustration of discrepancy in fatigue caused by an undocumented maintenance or rehabilitation event. Graphs. This figure contains six graphs that compare three discrepancies, which are Manual, PADIAS 1.X, and High-Low. Graph A. Discrepancy noted in fatigue cracking in survey of August 1996. The date is graphed on the horizontal axis from November 1989 to August 1996. Fatigue is graphed on the vertical axis from 0 to 200. In November 1989, High-Low tested 0 and 145, and PADIAS 1.X tested 65. In 1990, High-Low tested 0 and 70, and PADIAS 1.x tested 40. In 1995, High-Low tested 0 and 180, and Manual tested 90. Graph B. Discrepancy noted in longitudinal cracking in the wheel path in the survey of August 1996. The date is graphed on the horizontal axis from November 1989 to August 1996. Longitudinal Crack WP is graphed on the vertical axis from 0 to 400. In 1995, High-Low tested 10 and 300, and Manual tested 180. PADIAS 1.X tested zero in 1989 and 1990, and Manual tested zero in August 1996. Graph C. Discrepancy noted in non-wheel path longitudinal cracking in survey of August 1996. The date is graphed on the horizontal axis from November 1989 to August 1996. Longitudinal Crack NWP is graphed on the vertical axis from 0 to 500. In November 1989, High-Low tested 20 and 160, and PADIAS 1.X tested 95. In 1990, High-Low tested 90 and 340, and PADIAS 1.x tested 210. In 1995, High-Low tested 90 and 430, and Manual tested 280. Graph D. Discrepancies noted in number of transverse cracks in surveys of October1995 and August 1996. The date is graphed on the horizontal axis from November 1989 to August 1996. Transverse Crack Number is graphed on the vertical axis from 0 to 150. In November 1989, High-Low tested 35 and 60, and PADIAS 1.X tested 45. In 1990, High-Low tested 50 and 100, and PADIAS 1.X tested 75. In 1995, High-Low tested 25 and 40, and Manual tested 35. Graph E. Discrepancies noted in length of transverse cracks in surveys of October 1995 and August 1996. The date is graphed on the horizontal axis from November 1989 to August 1996. Transverse Crack Length is graphed on the vertical axis from 0 to 150. In November 1989, High-Low tested 40 and 60, and PADIAS 1.X tested 50. In 1990, High-Low tested 80 and 140, and PADIAS 1.X tested 110. In 1995, High-Low tested 25 and 45, and Manual tested 40. Graph F. Discrepancy noted in block cracking in survey of August 1996. The date is graphed on the horizontal axis from November 1989 to August 1996. Block is graphed on the vertical axis from 0 to 600. In 1995, High-Low tested 190 and 530, and Manual tested 375. PADIAS 1.X tested zero in 1989 and 1990, and Manual tested zero in August 1996. On August 1996, Manual tested zero for graphs A through F. A few months before August 1996, Manual and high-low tested a higher amount of cracks. There is a discrepancy in figure 15 shown by the difference tested on August 1996 and the years before August 1996.

Equations

Equation 1. Upper and Lower Limit equals X, which is the distress value of interest, plus or minus 3 times COV, which is the coefficient of variation, times X.

TFHRC Home | FHWA Home | Feedback United States Department of Transportation - Federal Highway Administration