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.
83 ME Design program in general have a reasonable capability to estimate the field transverse cracking performance of HMA and WMA pavements. Results from some local validation studies also indicate that thermal cracking prediction results obtained using local calibration factors are acceptable (Von Quintus et al. 2007, FHWA 2010). However, the Pavement ME Design program under- predicted the thermal cracking in the SC 178 pavement, and the prediction rankings of the MO Rte. CC and MO Hall St. pavements do not match the field measurements very well. All three of these projects contained jointed plain concrete pavement in their existing pavements. Therefore, the Pavement ME Design program may not be suitable for predicting ther- mal cracking in asphalt overlays that have been placed on top of an existing concrete pavement. The literature also suggests that using the Pavement ME Design program prediction results for such a scenario can be an improper approach (Zhou et al. 2013, Saxena et al. 2010, Von Quintus et al. 2012). In the Pavement ME Design program, 83.4 ft/mile and 2,592.7 ft/mile are the minimum and maximum prediction values, respectively, based on a 95 percent reliability level. Among the 48 HMA and WMA pavements evaluated, 33 pave- ments have predicted values close to 83.4 ft/mile and 10 have predicted values equal to 2,592.7 ft/mile, and only 5 sections have other values in between. Because most predicted ther- mal cracks either have no predicted values or reach maximum magnitudes, whereas field measurements were assigned dif- ferent severity levels, a weak correlation between the field and predicted values was observed. The reason that so many predictions are located at either the minimum or maximum extremes could be related to the two assumptions applied in the current thermal cracking prediction model, that is, TCModel (Hiltunen and Roque 1994): 1. A crack is not counted as a crack until the local vertical crack propagates through the entire depth of the asphaltic concrete surface layer. The data collected during this study have provided an oppor- tunity to evaluate the effectiveness of AASHTOWare Pave- ment ME Design software to predict the field performance of hot mix asphalt (HMA) and warm mix asphalt (WMA). AASHTOWare Pavement ME Design analysis was performed for each pavement using field-collected information and laboratory-tested material properties. Nationally calibrated default performance prediction models were used in the analyses. Level 1 material properties were used for overlays as much as possible, and a combination of Level 2 and Level 3 was used for existing asphalt concrete and existing Portland cement concrete (PCC) layers, as well as for base and subgrade layers. Other collected input data include pavement structure and layer thickness, traffic data such as average annual daily truck traffic (AADTT), climate, design speed, falling weight deflec- tometer test results, and back-calculated layer modulus values. Comparisons between the field-measured distresses and the predicted results derived from the Pavement ME Design soft- ware were analyzed. AASHTOWare Pavement ME Design Analysis Results for Transverse Cracking Figure E.1 (a) presents the predicted thermal cracking results and Figures E.1 (b) and (c) present the field-measured trans- verse cracking results for the 2012 first-round distress survey and 2014 second-round distress survey, respectively. These comparisons between the predicted results and the field mea- surements are based on 17 projects that have complete project information. A comparison between the predicted results and the field- measured distress values indicates that the field-measured trans- verse cracking results presented in Figures E.1 (b) and (c) exhibit a relatively consistent trend with the predicted thermal cracking values presented in Figure E.1 (a). This finding indicates that the thermal cracking models in the AASHTOWare Pavement A P P E N D I X E AASHTOWare Pavement ME Design Analysis Results
(a) 0 500 1000 1500 2000 2500 3000 Pr ed ic te d Th er m a l C ra ck in g (ft /m ile ) HMA Sasobit Evotherm Foaming Advera LEA OH MD MN CO WA PA WA NV IL SC MO MO PA TX TX CA CA 541 925 169 I-70 SR 12 2012 I-90 Bravo 147 178 CC Hall 2006 251 324 3a 3b (b) 0 50 100 150 200 250 300 350 1s t-R ou nd T ra ns ve rs e C ra ck in g (ft /20 0 f t) HMA Sasobit Evotherm Foaming Advera LEA OH MD MN CO WA PA WA NV IL SC MO MO PA TX TX CA CA 541 925 169 I-70 SR 12 2012 I-90 Bravo 147 178 CC Hall 2006 251 324 3a 3b (c) 0 50 100 150 200 250 300 350 2n d- R ou nd T ra ns ve rs e C ra ck in g (ft /20 0 f t) HMA Sasobit Evotherm Foaming Advera LEA OH MD MN CO WA PA WA NV IL SC MO MO PA TX TX CA CA 541 925 169 I-70 SR 12 2012 I-90 Bravo 147 178 CC Hall 2006 251 324 3a 3b Figure E.1. Comparisons of (a) Pavement ME Design program prediction results for thermal cracking, (b) first-round field transverse cracking, and (c) second-round field transverse cracking.
85 2. As long as the crack depth reaches the maximum value (the depth of the total pavement layers), the amount of thermal cracking will be maximized accordingly. The value will be a constant and will not further change. It is also worth noting that the predicted thermal cracking length values for the PA SR 2006 project are much higher than those for the PA SR 2012 project, although both projects share similar climate conditions. In the field, both projects were reported to have no transverse cracking. The possible reason for this difference between the results for these two projects is believed to be the lower creep compliance values for the overlay mixtures of the PA SR 2006 project compared with those obtained from the PA SR 2012 project. When the creep compliance values of the SR 2012 project were used as the SR 2006 project input, while the other properties of the SR 2006 project remained the same, minimal thermal crack- ing was predicted for the SR 2006 project. In other words, the current thermal cracking model is very sensitive to creep compliance, whereas field thermal cracking may not neces- sarily be so. AASHTOWare Pavement ME Analysis Results for Wheel-Path Longitudinal Cracking As shown in Figure E.2, the Pavement ME Design pro- gram predicted significant wheel-path longitudinal cracking for the MD 925, CO I-70, and CA 3b projects and insignifi- cant wheel-path longitudinal cracking for the other projects, whereas field longitudinal cracking was reported only for the OH SR 541, SC 178, MO Hall St., and TX 324 projects in the first-round distress survey. In the second-round survey, the crack lengths that were measured in 2012 had increased. Some wheel-path longitudinal cracks developed between 2012 and 2014 in the MD 925, CO I-70, WA I-90, and PA SR 2006 projects, although the crack severity for most of these projects was minor. When compared with the field measurements, the findings indicate that the Pavement ME Design program may not be able to predict longitudinal cracking comparable with that measured in the field. Some local validation studies also concluded that the HMA longitudinal cracking model must be refined in order to improve the accuracy of the predictions (Quintus et al. 2007, Ceylan et al. 2013). AASHTOWare Pavement ME Design Analysis Results for Rutting Performance Figure E.3 presents comparisons between the predicted rut depths and the second-round survey field rut depths. The predicted results for most projects match those measured in the field fairly well, except that the Pavement ME Design pro- gram generally over-predicted the field rut depth values by 30 percent to 40 percent. This finding indicates that the rut depth models in the Pavement ME Design program in gen- eral have a reasonable capability to predict the field rut depths of HMA and WMA pavements.
86 (a) 0 1000 2000 3000 4000 5000 6000 7000 8000 Pr ed ic te d To p- do w n Fa tig ue C ra ck in g (ft /m ile ) HMA Sasobit Evotherm Foaming Advera LEA OH MD MN CO WA PA WA NV IL SC MO MO PA TX TX CA CA 541 925 169 I-70 SR 12 2012 I-90 Bravo 147 178 CC Hall 2006 251 324 3a 3b (b) 0 100 200 300 400 500 600 700 800 1s t-R ou nd F ie ld L on gi tu di na l C ra ck in g (ft /20 0 f t) HMA Sasobit Evotherm Foaming Advera LEA OH MD MN CO WA PA WA NV IL SC MO MO PA TX TX CA CA 541 925 169 I-70 SR 12 2012 I-90 Bravo 147 178 CC Hall 2006 251 324 3a 3b 0 100 200 300 400 500 600 700 800 2n d- R ou nd F ie ld L on gi tu di na l C ra ck in g (f t/2 00 ft ) HMA Sasobit Evotherm Foaming Advera LEA OH MD MN CO WA PA WA NV IL SC MO MO PA TX TX CA CA 541 925 169 I-70 SR 12 2012 I-90 Bravo 147 178 CC Hall 2006 251 324 3a 3b (c) Figure E.2. Comparisons of (a) Pavement ME Design program prediction results for top-down cracking, (b) first-round field longitudinal cracking, and (c) second-round field longitudinal cracking.
87 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Pr ed ic te d R ut D ep th o f H M A (in .) HMA Sasobit Evotherm Foaming Advera LEA OH MD MN CO WA PA WA NV IL SC MO MO PA TX TX CA CA 541 925 169 I-70 SR 12 2012 I-90 Bravo 147 178 CC Hall 2006 251 324 3a 3b (a) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Fi el d R ut D ep th (i n.) HMA Sasobit Evotherm Foaming Advera LEA OH MD MN CO WA PA WA NV IL SC MO MO PA TX TX CA CA 541 925 169 I-70 SR 12 2012 I-90 Bravo 147 178 CC Hall 2006 251 324 3a 3b (b) Figure E.3. Comparisons between (a) Pavement ME Design program prediction results for rut depths of HMA layers and (b) second-round survey field rut depths.
