Investigating the Relationship of As-Constructed Asphalt Pavement Air Voids to Pavement Performance (2021)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

147 Background The NCHRP 20-50(18) study objective was to determine the effect of in-place air voids on the performance of asphalt concrete (AC) pavements. The research focused on four primary distress types related to asphalt pavement performance: rutting, fatigue cracking, thermal cracking, and ride. The research team did not examine a composite pavement performance index because there was no single accepted index used by all agencies. The research team applied three analysis methods, described as Analysis Method 1, 2 and 3, to LTPP data and then validated the results with data from other sources. Although there are other pavement performance characteristics, such as raveling, they are not the focus of this study. Further, this study does not include composite pavements. The terms “in-place air voids” and “as-constructed air voids” are often used interchangeably. In-place air voids can be defined as the asphalt mixture air voids as they change over time, which an agency does not routinely measure and cannot control. As-constructed air voids (AV) is a measured value that an agency can control in their construction specifications. The team applied the as-constructed AV value that best corresponds to the performance characteristic being analyzed. This approach required the input data for each pavement section in the study to have construction records for each AC lift. The research team recognized that as- constructed AV was not the only material parameter that influences pavement performance. In fact, other material properties play a more significant role but are not the objective of this study. They are all consolidated into the pavement structure input values. The primary source of input and output dataset for the study was the LTPP database. The LTPP database contains an extensive amount of detail for each section. The research team selected climate, traffic, and pavement structure input variables and measured rutting, cracking, and ride performance data as output variables. Time-based input values, such as climate and traffic, are consolidated into annual values. Material input values, specifically the pavement structure layers, are separated into layer thickness and stiffness values. Some assumptions were made regarding the nature of the measured performance distress. The measured rutting was not examined to establish the cause or links to associated pavement material properties or the pavement structure. Fatigue performance was measured as wheel path cracking and was not examined to distinguish between the causes of the cracking, such as bottom-up, top-down, or delamination (commonly caused by poor construction). Similarly, thermal cracking was measured as any transverse crack. For the analysis of rehabilitation sections, both wheel path and transverse cracking are assumed to be reflective cracking. Ride was examined because it was a common performance characteristic measured by highway agencies, but the measured roughness was not examined to establish the cause. The ride performance most likely mirrors other distress in the pavement. Table 1 describes the input and output dataset used for the study.

148 Table 1. Input and Output Dataset Summary Data Category Data Type and Description AM-1(1) Req’d AM-2(1) Req’d AM-3(1) Req’d Climate Analysis Method 1 categorized climate by four regions (dry+freeze, dry+no freeze, wet+freeze, and wet+no freeze), similar to the LTPP climate regions. Analysis Methods 2 and 3 assembled annual values for temperature, precipitation, and freezing index over the period of performance. 25-year average annual temperature, computed from annual values X 25-year average annual precipitation, computed from annual values X Average Annual Temperature (Deg C), converted to Deg F (CLM_VWS_TEMP_ANNUAL.MEAN_ANN_TEMP_AVG) X X Annual Freezing Index (Deg C Degree Days) , converted to Deg F Degree Days (CLM_VWS_TEMP_ANNUAL.FREEZE_INDEX_YR) X X Total Annual Precipitation (mm), converted to inches (CLM_VWS_PRECIP_ANNUAL.TOTAL_ANN_PRECIP) X X Traffic Traffic was categorized by equivalent single axle loads (ESAL). Analysis Method 1 used cumulative 5- year values for purposes of dividing the pavement sections into traffic load categories. Analysis Methods 2 and 3 assembled the sequence of total annual ESALs over the period of performance. Cumulative 5-year ESALs, computed for the study from annual values X Yearly Computed ESALs (TRF_ESAL_COMPUTED.KESAL_YEAR) X X Pavement Pavement structure was a complicated input variable due to the differences in available materials and in design methods used by highway agencies. Analysis Method 1 consolidated the entire pavement structure to an AASHTO structural number (SN) value for purposes of dividing the pavement sections into pavement structure groups. Analysis Methods 2 and 3 used pavement layer thicknesses for asphalt and base and used layer stiffness modulus values for each layer. Structural Number (TRF_ESAL_INPUTS_SUMMARY.SN_VALUE) X Asphalt and Base Layers Thickness X X Asphalt Layer Dynamic Modulus (E*) measured at 14F & 0.5 Hz, 70F & 10 Hz, and 130F & 10 Hz. When measured E* values were not available, E* was estimated from five hierarchical models. X X Unbound Base Layer Resilient Modulus X X Modulus of Subgrade Reaction (INV_SUBGRADE.SUBGRADE_REACTION_MODULUS) X X Age (yrs) (based on date of construction or open to traffic) X X X New Construction Versus Rehabilitation X X X As- Constructed Air Voids As-constructed air voids (AV) was a complicated variable to quantify. An asphalt pavement structure is generally paved in multiple lifts and the lifts may not have the same AC mixtures. The team applied the as-constructed AV value that best corresponded to the performance characteristic being analyzed. For Analysis Methods 2 and 3, the analysis of rutting, ride, and thermal cracking used the as-constructed AV value of the surface asphalt lift and fatigue cracking applied a weighted-average as-constructed AV value for the entire asphalt layer. This approach required the input data for each pavement section to have records for each AC lift. Surface Lift (based on measures reported in the LTPP database): Could be cores (within 6 months of construction), DOT QC X X X Asphalt Layer Average, based on weighted values for each layer thickness X X X Performance See text in cells below for a discussion of data assembled for each performance characteristic. Rutting (in) based on maximum rutting measured in either wheel path over the 500-ft LTPP section length X X X Fatigue Cracking (% of lane width over the section length) based on wheel path cracking expressed as percent cracking of the 39-inch wheel path based on total lane width. The maximum value is 50 to 60% based on lane width. This value was selected because it was required by FHWA’s HPMS reporting. X X X Thermal Cracking (ft/mi) based on total measured length of transverse cracks over the 500-ft LTPP section length X X X Ride (in/mi) based on mean value of international roughness index (IRI) in both wheel paths using multiple measured passes over the 500-ft LTPP section length X X X Note 1: AM-1 = Analysis Method 1, AM-2 = Analysis Method 2, AM-3 = Analysis Method 3 Summary statistics of the test section input variables for the 422 LTPP pavement sections used in the study are presented in Table 2. Note that the reported minimum and maximum values for some inputs represent extreme values that would not be considered normal. Obvious examples are 221 inches of base/subbase and 4400 ksi for asphalt mixture E* at 130°F. For this

149 reason, the User Guide recommends examining only pavement sections with input variables that range between the 5th percentile (p5%) and 95th percentile (p95%) values. Table 2. Summary Statistics of Pavement Section Input Datasets Variable Units Mean Standard Deviation Minimum p05% p50% p95% Maximum AV_s %Gmm 6.1 3.0 0.1 1.9 5.8 11.2 21.8 AV_m %Gmm 6.0 2.4 1.4 2.6 5.8 10.4 15.2 Th_AC in 7.2 2.8 1.4 3.9 6.9 12.8 21.9 Th_BSB in 18.4 16.5 0.0 4.2 14.9 42.8 221.5 OvTh_ACa in 3.6 1.7 0.9 1.5 3.3 6.8 9.2 E*_IRI ksi 1,082.4 456.3 406.0 575.0 975.1 1,781.2 4,398.9 E*_FtC ksi 1,074.2 411.7 225.0 575.0 1,046.1 1,688.6 4,398.9 E*_TrC ksi 3,167.8 644.8 227.6 2,369.5 3,196.2 4,058.1 5,599.0 E*_Rut ksi 99.2 216.2 16.9 43.5 73.7 156.6 4,398.9 Mr_BSB ksi 32.3 26.8 9.1 14.7 21.1 100.0 100.0 Mr_SG ksi 9.1 3.0 3.3 4.9 8.8 15.1 17.9 BC_s % 5.0 0.7 3.6 4.0 4.9 6.4 9.0 BC_m % 4.8 0.7 3.5 3.8 4.8 6.0 9.0 Gr_4 % 55.9 10.1 22.0 39.0 55.0 72.9 90.0 Gr_200 % 6.1 1.6 0.9 3.7 6.4 8.9 13.0 Gr_k - 5.3 2.2 2.2 3.0 5.1 8.1 15.3 Gr_lambda - 0.8 0.2 0.6 0.7 0.8 1.1 1.7 aStatistics correspond to the positive AC overlay total thickness values Three approaches were used to analyze the LTPP dataset: common subsets, regression, and artificial neural networks. All three analysis methods examined the new construction and pavement rehabilitation LTPP sections as separate groups. Analysis Method-1 created common subsets of input data based on climate, traffic, and pavement structure, then plotted a performance characteristic against as-constructed AV to establish a scatter plot representing the influence of as-constructed AV on pavement performance. Climate input was based on a modified version of the LTPP’s four climate zones. Traffic input was sorted into two categories based on five-year cumulative ESALs from the time the surface was open to traffic. Pavement structure was quantified by the composite SN value of the base and asphalt pavement layers as reported in the LTPP database. The rutting performance characteristic was measured rutting, in inches, after four years of traffic on the asphalt surface. The fatigue cracking performance characteristic was percent cracking in the 39-inch wheel path after 10 years of traffic. The thermal cracking performance characteristics were the year that transverse cracking initiated and the measured transverse cracking expressed as lineal feet per mile after three years from the time cracking initiated. The ride performance characteristics were the mean international roughness index, in inches per mile, after 10 years of traffic on the surface and the number of years for the initial ride value to increase 25 in/mi. Analysis Method-2 correlated (regressed) a broad array of climate, traffic, and pavement input values against the recorded LTPP performance over time. The final regression equation for each performance characteristic expresses the most significant variables influencing that specific pavement performance characteristic. As-constructed AV are always included in the regression regardless of its degree of significance. With the input values for a specific pavement, Analysis-2 generates a family of curves showing the incremental influence of as-constructed AV on the selected performance characteristic (y-axis) over time (x-axis).