Consideration of Preservation in Pavement Design and Analysis Procedures (2015)

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.

36 C H A P T E R 6 The application of preservation treatments could result in changes in pavement material properties (e.g., modulus), pave- ment structural properties (e.g., thickness, moisture content), moisture and thermal profiles in the pavement system, and the level and rate of distress and roughness development over time. These changes will influence pavement performance and life. Therefore, by identifying the MEPDG inputs or model- ing components that are affected by the treatment application, quantifying the changes attributed to treatment application, and using the adjusted values of these items in the MEPDG design analysis process, the effect of preservation treatment on pavement performance and life will be accounted for. Process Description The application of a preservation treatment can result in changes in distress/roughness, material properties, structure cross-sections, and moisture and thermal profiles. Also, some preservation treatments can alter the distress/roughness level immediately upon application (e.g., the application of a thin HMA overlay would eliminate cracking, reduce the depth of rutting, and decrease IRI) and the rate and level of distress/ roughness redevelopment. Therefore, it will be necessary to define the adjustment that should be made to each MEPDG performance parameter, recognizing the following: • The empirical reflection cracking model may be used to predict the percentage of cracks (fatigue and thermal) or joints that propagate through the preservation treatment over time. This model uses a sigmoidal function with a and b fitting parameters that are a function of overlay (in this case, treatment) thickness, as well as c and d user-defined cracking progression parameters. • A dynamic rutting model that uses the base rutting model and a subtraction term that represents the change in rut- ting due to the application of a preservation treatment may be used. As illustrated in Figure 13, for every preservation treatment that is applied to reduce rutting (to zero), a reduc- tion of rutting by 0.25 in. is factored into the base model. Thus any rutting that occurs after treatment application is modeled as “base model rut depth minus 0.25 in.” If a sec- ond treatment is applied, then the rutting that takes place after the second treatment is modeled as “base model rut depth minus 0.25 in. minus 0.25 in.” • A dynamic faulting model can be created and used in a manner similar to rutting. • No adjustments to the overlay smoothness models are needed. The initial IRI in these models will be the value immediately upon preservation treatment application (as specified by the user). The other terms in the models (crack- ing and rutting) will be derived from their respective models. The treatment application can have either an immediate or long-term effect on the properties of the surface layer material of the pavement. For example, applying a fog seal or rejuvena- tor to an HMA pavement or performing a surface recycle will immediately soften the HMA surface and lead to a reduced modulus value and influence flexibility and resistance to load and environment initially and over time. To properly account for the effects of the changes in material properties on perfor- mance, these changes must be quantified. Some preservation treatments may not have an immedi- ate effect on the properties of the surface layer material but may influence the long-term properties of that material. For example, placing a surface treatment on an HMA pavement protects the HMA surface layer from ultraviolet (UV) expo- sure, thus reducing the rate at which the binder in the sur- face layer hardens with time (i.e., protects against aging). The aging model in the MEPDG includes both a surface aging model and a viscosity-depth model for predicting binder vis- cosity at any time and any depth in the pavement structure. A preservation treatment can also result in a change in the pavement structure cross-section. The thickness of the pave- ment surface layer may be reduced, as in the case of milling Using Modified Material and Pavement Structural Properties in MEPDG Models to Account for Preservation

37 of HMA or diamond grinding of PCC, or increased, as in the case of applying a surface treatment or thin HMA overlay. Although some treatments with large thicknesses are applied and treatments that are placed repeatedly over time could increase structural capacity, pavement preservation treatments are generally considered to have no effect on structural capac- ity. To model material characteristics and moisture and tem- perature regimes in the pavement structure, it is necessary to define the thickness and mechanistic properties associated with each preservation treatment. The preservation treatment can also influence the moisture and thermal profiles of the pavement system over time, as modeled by the Enhanced Integrated Climate Model (EICM). Treatments that seal or waterproof a pavement may reduce the infiltration of surface water into the structure and foun- dation, thereby reducing the moisture content and increasing the resilient modulus of the underlying unbound materials. Similarly, thick treatments may influence the thermal char- acteristics throughout the pavement. To capture these effects, certain treated pavement structure inputs, such as the infiltra- tion potential of the pavement (surface layer[s] and treated base layer[s]), the cross-slope and drainage path length of the treated pavement surface, and the surface shortwave absorp- tivity of the treatment, would need to be redefined. Tables 20 and 21 list the likely effects of different preser- vation treatments on performance indicators for HMA- and PCC-surfaced pavements, respectively. The process for determining the changes in material proper- ties resulting from the application of preservation treatments and their effect on pavement performance is summarized as follows. 1. Identify the Basic Pavement Structure and the Preservation Treatment Type: The original/base design and correspond- ing use scenario are identified, together with the specific preservation treatment that will be considered for applica- tion at some time following construction. 2. Identify Preservation Treatment Timing: The timing for the preservation treatment application is identified based on specific schedule or thresholds for performance indicators (e.g., the amount of transverse cracking or rutting). 3. Identify Baseline Material Properties of Pavement Structure and Treatment: Key material properties of the base pave- ment structure and the preservation treatment (such as engi- neering and thermal properties [e.g., dynamic modulus, creep compliance, coefficient of thermal contraction] and volumetric properties [e.g., air voids, mix density, effective asphalt content]) are identified. Tables 22 through 24 list the specific preservation treatment material inputs. Several preservation treatments reduce or delay the infil- tration of moisture through existing surface cracks and joints and may therefore increase the resilient modulus of the unbound and subgrade layers. However, this increase in stiffness will diminish over time. The resilient modulus for the unbound and subgrade layers may be determined from NDT (e.g., FWD backcalculation) or correlations with other tests (e.g., California bearing ratio [CBR] and R-value), or using values (AASHTO 2008). 4. Quantify Treatment Effect on Pavement Thickness: The effect of the preservation treatment on the existing pavement structure is quantified in terms of reduced or added struc- ture thickness. For example, chip seals, microsurfacing, and overlays will add a layer to the pavement structure, but milling and diamond grinding will reduce the surface layer thickness. 5. Identify Treatment Effect on Existing Layer Material Prop- erties and on Moisture and Thermal Properties of Pavement Structure: Short- and long-term effects of the preservation Base Model Rut = f(N) Load Cycles (N) Rut Depth, in. Rut Threshold = 0.25 in. Treatment 1 Immediate Adjustment = 0.25 in. Treatment 2 Immediate Adjustment = 0.25 in. Dynamic Model Rut = f(N) – 0.25 in. Dynamic Model Rut = f(N) – 0.25 in. – 0.25 in. Figure 13. Concept of adjusting rutting model to account for preservation treatment effects.

38 Treatment Performance Indicator Total Rutting (HMA and unbound) Transverse Thermal Cracking Fatigue Cracking (Bottom-up Alligator) Fatigue Cracking (Top-Down Longitudinal) Reflection Cracking (in overlays) Smoothness (IRI) Crack Filling/Sealing (+)/+ (+)/+ Fog Seal/Rejuvenator Seal Sand/Scrub Seal Slurry Seal (+) (+) Microsurfacing + (+) (+) + Chip Seals (+) (+) (+) (+) Thin HMA Overlays (+) (+) (+) + Ultrathin HMA Overlays (+) (+) (+) (+) Ultrathin Bonded Wearing Course (+) (+) (+) (+) Hot In-Place Recycling + + + + + + Cold In-Place Recycling + + + + + + Ultrathin Concrete Overlay + + + + + + Notes: + or − = significant or long-term positive or negative impact; (+) or (−) = moderate or short-term positive or negative impact; = slight positive impact; = slight negative impact; blank cells designate no effect. Table 20. Possible effects of preservation treatments on performance indicators of HMA-surfaced pavements. Treatment Performance Indicator JPC Pavement CRC Pavement Smoothness (IRI) Crack/ Joint Faulting Load Transfer Efficiency Trans- verse Cracking Crack/ Joint Spalling Crack Spacing/ Width Load Transfer Efficiency Punchouts Crack Sealing/ Joint Resealing + Diamond Grinding + + Diamond Grooving Partial-Depth Repair + Full-Depth Repair + + + + + + + Load Transfer Restoration + + (+) Cross-Stitching + + (+) Thin HMA Overlay (+) + Ultrathin Bonded Wearing Course (+) (+) Notes: + or − = significant or long-term positive or negative impact; (+) or (−) = moderate or short-term positive or negative impact; = slight positive impact; = slight negative impact; blank cells designate no effect. Table 21. Possible effects of preservation treatments on performance indicators of PCC-surfaced pavements. Input Level 1 Level 2 Level 3 Superpave Performance Grade Binder AASHTO T 49 Same as Level 1 Superpave performance grade Penetration/Viscosity Grade Binder AASHTO T 49, T 53, T202, T 201, T 228, and TP 85 Same as Level 1 Penetration/viscosity grade Table 22. Summary of asphalt binder material inputs (AASHTO 2008, Pierce et al. 2010).

39 Input Level 1 Level 2 Level 3 Unit Weight AASHTO T 166 Not applicable Typical value (default = 150 lb/ft3) Effective Binder Content AASHTO T 308 Not applicable Typical value (default = 11.6%) Air Voids AASHTO T 166 Not applicable Typical value (default = 7%) Poisson’s Ratio Not applicable Reference temperature Typical value (default = 0.35) Dynamic Modulus AASHTO TP 62 Binder properties and aggregate gradation Same as Level 2 Indirect Tensile Strength AASHTO T 322 Same as Level 1 Calculated internally Creep Compliance AASHTO T 322 at -4, 14, and 32 F AASHTO T 322 at 14 F Calculated internally Thermal Conductivity Not applicable Not applicable Typical value (default = 0.67 BTU/ft-hr- F) Heat Capacity Not applicable Not applicable Typical value (default = 0.23 BTU/lb- F) Thermal Contraction Not applicable Mix and aggregate Calculated internally Table 23. Summary of HMA material inputs (AASHTO 2008, Pierce et al. 2010). Input Level 1 Level 2 Level 3 Unit Weight AASHTO T 121 Not applicable Typical value (default = 150 lb/ft3) Poisson’s Ratio ASTM C469 Not applicable Typical value (default = 0.20) Coefficient of Thermal Expansion AASHTO TP 60 Not applicable Typical value (default = 5.5 x 10-6 in./in./°F) Thermal Conductivity ASTM E1952 Not applicable Typical value (default = 1.25 BTU/ft-hr-°F) Heat Capacity ASTM D2766 Not applicable Typical value (default = 0.28 BTU/lb-°F) PCC Set Temperature Not applicable Not applicable Internally calculated or user-defined Ultimate Shrinkage Not applicable Not applicable Internally calculated or user-defined Reversible Shrinkage Not applicable Not applicable User-defined PCC Strength AASHTO T 97, ASTM C469 AASHTO T 22 User-defined Table 24. Summary of PCC material inputs (AASHTO 2008, Pierce et al. 2010). on the performance of the existing pavement is determined (e.g., reducing rutting to zero or IRI value to a certain level). 7. Establish MEPDG Reflection Cracking Model Coefficient and Dynamic Models for Rutting/Faulting: The MEPDG reflection cracking model coefficient d, which governs the acceleration (d > 1) or delay (d < 1) in the forma- tion of reflective cracks (from fatigue and transverse cracks in existing HMA pavement) in the preservation treatment, is determined. Also, a rut depth (or faulting for PCC pavement) model is proposed that modifies the MEPDG base model to account for the immediate change in rut depth (or faulting) by including an adjust- ment term. treatment on the existing surface layer material proper- ties (i.e., changes in engineering or thermal properties, or volumetric properties of the HMA surface layer), on the moisture and thermal profiles of the pavement struc- ture (e.g., drainage/infiltration potential, cross-slope, and drainage path length) are identified. However, the MEPDG considers only the effects of shoulder type, edge drains, and drainage layers (AASHTO 2008); it allows changes to the layer moduli of the unbound and subgrade layers and the surface shortwave absorptivity but not to the infiltra- tion rate. These effects should be defined and considered. 6. Identify Immediate Treatment Effect on Performance of Pavement Structure: The immediate effect of the treatment

40 The empirical reflection cracking model can be used to predict the percentage of cracks (fatigue and thermal) or joints that propagate through the preservation treatment over time. The MEPDG user-defined cracking progression parameters c and d can be adjusted to account for delay- ing or accelerating the progression of reflection cracking. The MEPDG Manual of Practice (AASHTO 2008) provides recommended values for c and d, but other values’ param- eters should be determined from calibration. Because the d parameter depends on overlay thickness and does not easily distinguish between fatigue and reflection cracking in the overlay, reliability of the reflection cracking model is set at 50% and cannot be changed by the user. A dynamic rut depth model that uses the MEPDG base rut depth model and a subtraction term can be developed to consider the immediate rut depth change due to pres- ervation treatment application. Figure 14 illustrates two preservation treatment applications, each of which reduces the rut depth to zero when the threshold value of 0.25 in. is reached. The dynamic rut depth model applies an imme- diate adjustment of 0.25 in., after which rut depth pro- gresses as defined by the base model. A dynamic faulting model can be developed in a manner similar to that described for the rut depth. The concept is illustrated in Figure 15. No adjustments to the overlay smoothness models are required. The initial IRI in these models will be the value specified as an immediate adjustment corresponding to the preservation treatment. Table 25 lists the effects of various preservation treatments on IRI as reported in the literature. 8. Perform Pavement ME Design Analysis: The base design is analyzed using design inputs for traffic, climate, and M ea n Jo in t Fa ul ti ng (i n. ) Pavement Age (years) Treatment 1 Adjustment = 0.12 in. "Base" Model Threshold = 0.12 in. "Dynamic" Model Treatment 2 Adjustment = 0.12 in. Figure 15. Concept of dynamic faulting model. To ta l R ut D ep th (i n. ) Pavement Age (years) Treatment 1 Adjustment = 0.25 in. "Base" Model Threshold = 0.25 in. "Dynamic" Model Treatment 2 Adjustment = 0.25 in. Figure 14. Concept of dynamic rut depth model.

41 materials properties, a specific design life, reliability lev- els for the individual performance indicators, and perfor- mance indicator threshold values for rehabilitation. Either the MEPDG performance prediction models or locally calibrated models can be used. 9. Perform Pavement ME Design Analysis for Preservation- Treated Design: A design analysis similar to that performed for the base design is performed for the preservation-treated design using the same base design parameters to consider the effects of the preservation treatment. The output from the base design (i.e., predicted distress and roughness levels) covering the period from original construction to the time when the first performance indicator threshold is reached is combined with the output from the preservation-treated design to produce the output for the specified design life. The effects of the treatment can then be evaluated in terms of (a) the immediate change in distress/roughness and their redevelopment, (b) the immediate or long-term change in the mechanistic properties of the pavement surface layer, (c) the immediate change in the pavement structural cross- section, and (d) the change in the moisture or thermal properties of the pavement surface layer and their effect on moisture or temperature profiles throughout the pave- ment structure. Table 26 lists the data elements required for the design analysis of the baseline/untreated and preservation-treated alternatives. Feasibility Assessment Modifying material properties involves defining the types of effects of the application of a preservation treatment on a pavement (e.g., immediate and long-term changes in distress/ roughness levels, material properties of the surface layer of the pavement, pavement structure cross-section, and moisture and thermal profiles of the pavement system). The design analysis uses the Pavement ME Design software to develop predicted distress/roughness values for a base design and a correspond- ing preservation-treated design, and then merges the two sets of predictions. This approach addresses only the cracking, rut- ting, faulting, and smoothness models included in the MEPDG. The level of effort required to implement this approach is fairly significant. Although some of the required inputs (e.g., typical treatment types and applications, distress/roughness threshold levels for preservation and rehabilitation treatments) can be easily obtained, other inputs must be obtained through col- lection and analysis of actual data. Examples of these inputs include the rate of redevelopment of distress/roughness, the change in the HMA surface layer dynamic modulus, and the change in pavement layer drainage and moisture characteris- tics following preservation. A major drawback to this approach is the complexity of accurately defining the changes in prop- erties resulting from the application of different preservation treatments at different times during the life of the pavement. This approach requires no modifications to the Pavement ME Design software and entails no added complexity in the use of the program. It simply involves design analysis compu- tations for the original/base design, then performs the design analysis computations, repeats the process for the preservation- treated design, and merges the two sets of design outputs. Examples of Implementation Process Two hypothetical examples are presented to illustrate how modifying material properties could be used to account for preservation effects on performance. In one example, micro- surfacing is applied to an existing HMA-surfaced pavement, and in another example, diamond grinding is performed on an existing PCC-surfaced pavement. When possible, actual inputs have been included and all assumptions have been clearly stated. These examples use inputs obtained from the Colorado DOT (CDOT) Pavement Design Manual (CDOT 2013) and Standard Specifications for Road and Bridge Con- struction (CDOT 2011). In these examples, “default” refers Treatment Type Before Treatment IRI, in./mi1 After Treatment IRI, in./mi1 Percent Improvement Reference Diamond Grinding 130 256 57 168 56 35 Battaglia 2010 Pierce and Muench 2009 Hot In-Place Recycling 109 78 29 Browning 1999 Microsurfacing 92 77 15 Ji et al. 2011 Milling (2) (2) 6 West et al. 2011 Thin HMA Overlay (2) (2) 18 to 36 Labi et al. 2005 Ultrathin HMA Overlay 162 154 99 89 39 42 Hanson 2001 Corley-Lay and Mastin 2007 Notes: 1 Values shown are based on the average IRI of individual projects reported in the reference publications. Actual IRI improvement may vary and depends on the IRI value prior to treatment application and agency design and construction practices. 2 Values were not provided. Table 25. Reported effects of preservation treatments on IRI.

42 to the default values provided in the Pavement ME Design software. Example 1: HMA Pavement Preservation Step 1: Identify Baseline Pavement Design and Preservation Treatments The specifics of the baseline pavement design are: • Pavement type: Conventional flexible pavement • Design period: 20 years • Functional class: Principal arterial • Traffic: – Truck traffic classification (TTC): Predominantly single- trailer trucks (TTC 1) – Two-way average annual daily truck traffic (AADTT): 450 (assumed) – Number of lanes in the design direction: two – Percent trucks in design direction: 50 – Percent trucks in design lane: 95 – Vehicle class distribution and growth: Default – Monthly adjustment: Default – Axles per truck: Default – Operational speed: 50 mi/hr – Axle distribution: Default – Axle configuration: Default – Lateral wander: Default – Wheelbase: Default • Closest weather station: Cortez, CO Table 27 lists the CDOT-recommended preservation treat- ments for HMA-surfaced pavements (CDOT 2013). Step 2: Identify Preservation Treatment Timing It is assumed that microsurfacing will be applied 10 years after original construction. Step 3: Identify Baseline and Preservation Treatment Material Properties The following material properties for the baseline pavement are based on CDOT’s Standard Specifications for Road and Bridge Construction and Pavement Design Manual: • HMA: Grading SX (CDOT designation) – Mixture volumetrics Data Category Data Element Analysis Parameters • • • • • • • • • • • • • • • • Untreated design strategy—typical pavement design Preservation-treated design strategy—same typical pavement design, except with a specific preservation treatment included Design life Performance Criteria and Reliability HMA performance indicators—rut depth, reflection cracking, and IRI PCC performance indicators—faulting and IRI Design reliability (for individual distresses and smoothness) Structure Properties Untreated design strategy—layer types, materials, and thicknesses Preservation-treated design strategy—same as untreated Surface shortwave absorptivity Preservation Treatment Application Parameters Treatment timing – Distress, smoothness, and/or overall condition levels of original pavement at time of treatment application Existing HMA layer material properties Treatment effect on existing pavement structure – Removal depth of existing HMA surface (milling) – Treatment application thickness Treatment effect (short- and long-term) on existing HMA surface layer material properties – Dynamic modulus Treatment impact (short- and long-term) on moisture and thermal profile of existing pavement – Surface shortwave absorptivity – Unbound layer modulus Performance Modeling Parameters Immediate adjustment of post-treatment performance levels – Post-treatment distress/smoothness measurements Long-term adjustment of post-treatment distress level via rate of redevelopment of distresses/smoothness – Reflection cracking (for HMA-surfaced treatments) – Faulting (for PCC-surfaced treatments) Table 26. Data elements required for AASHTOWare Pavement ME Design analysis.

43 77 Unit weight: 150 lb/ft3 (default) 77 Effective binder content: 10% 77 Air voids: 4% 77 Poisson’s ratio: 0.35 (default) – Mechanical properties 77 Dynamic modulus: Level 3 77 Gradation: 100% passing ¾ in., 95% passing ³⁄8 in., 65% passing No. 4, and 6% passing No. 200 77 Reference temperature: 70°F 77 Asphalt binder type: PG 64-28 77 Indirect tensile strength: 464.65 lb/in.2 (internally calculated) 77 Creep compliance: Level 3 – Thermal properties 77 Thermal conductivity: 0.67 BTU/hr-ft-°F (default) 77 Heat capacity: 0.23 BTU/lb-°F (default) 77 Thermal contraction: 1.185 × 10–5 (internally calcu- lated) – Surface shortwave absorptivity: 0.85 (default) – Endurance limit: Not applied (not recommended until calibrated) – Layer interface: Full friction • Unbound base: Class 6 – Aggregate type: Crushed stone – Poisson’s ratio: 0.40 – Coefficient of lateral earth pressure: 0.5 (default) – Resilient modulus: 38,721 lb/in.2 (CDOT median value) – Gradation (median of specification range): 100% pass- ing ¾ in., 47.5% passing No. 4, 40% passing No. 8, and 7.5% passing No. 200 – Liquid limit: 10 – Plasticity index: 2 • Subgrade: A-2-6 – Poisson’s ratio: 0.40 – Coefficient of lateral earth pressure: 0.50 (default) – Resilient modulus: 16,000 lb/in.2 (default) – Gradation: Default – Liquid limit: 15 – Plasticity index: 5 Because the MEPDG and the Pavement ME Design software do not provide material inputs (e.g., dynamic modulus, indirect tensile strength [IDT], heat capacity) for microsurfacing, the microsurfacing material properties were assumed to be similar to those for HMA layers. Also, because microsurfacing could reduce the potential for moisture intrusion through any exist- ing cracks, an increase of 5% was assumed for the resilient mod- ulus of the base course and subgrade. (Actual changes would need to be quantified from in-service and laboratory testing.) Step 4: Quantify Effect of Treatment Application on Pavement Thickness Although the typical thickness of microsurfacing is 0.40 to 0.50 in., the minimum thickness of an overlay that can be con- sidered in the Pavement ME Design software is 1 in. Therefore, the microsurfacing thickness was assumed to be 1 in. Step 5: Identify Effect of Treatment Application on Existing Layer Material Properties Microsurfacing will be analyzed as an additional thickness of HMA; no modification to the existing asphalt concrete (AC) material properties will be required. Treatment Distress Types Addressed Typical Thickness Comments Crack Sealing High-severity linear cracks Not applicable — Patching Medium- to high-severity alligator cracking Varies depending on depth of distress — Chip Seal Cracking, surface aging Varies depending on aggregate size and number of applications Estimated performance life is 8 to 10 years. Thin Overlay or Microsurfacing Surface friction, hydroplaning, raveling, low-severity cracking, bleeding 0.4 to 0.5 in. Estimated performance life is 4 to 7 years. Leveling Course or Milling Rutting Varies depending on rut depth — Cold In-Place Recycling (w/HMA overlay) Not specified 2 to 4 in. Estimated performance life is 6 to 21 years. Hot In-Place Recycling (w/HMA overlay) Rutting, wearing, raveling, non- structural surface cracking, aging, poor frictional characteristics <2 in. Estimated performance life is 6 to 23 years. Table 27. Recommended preservation treatments for HMA-surfaced pavements (CDOT 2013).

44 Step 6: Identify Immediate Effect of Treatment Application on Existing Condition It is assumed that the application of the microsurfacing will reduce the rut depth to zero and IRI to 90 in./mi. Step 7: Determine Dynamic Model The dynamic model will assume reductions of the rut depth to zero (see Figure 14) and the IRI to 90 in./mi with the appli- cation of the microsurfacing layer. Step 8: Develop a Baseline Design The material inputs defined for the project (see Table 28) were entered into the Pavement ME Design program. The analysis determined that a 15-in.-thick pavement (7-in. HMA grading SX [PG 58-28] plus 8-in. Class 6 aggregate base) will meet all of the performance criteria (HMA layer thickness rounded up to the nearest 0.5 in.). The results of this analysis are listed in Table 29, and plots for IRI, rut depth, thermal cracking, and fatigue cracking (corresponding to 90% reliability) over time are shown in Figures 16 through 19, respectively. As seen in these fig- ures, the critical distress for the baseline design is HMA rutting (i.e., HMA rut depth reaches the threshold value of 0.25 in. by the end of the 20-year design period), at which time a preservation treatment may be applied to reduce future rutting. Step 9: Develop a Preservation-Treated Design The MEPDG and the Pavement ME Design software can be used to estimate the change in the performance or pave- ment life due to the application of a preservation treatment (Figure 20) or determine the required baseline design thick- ness if a preservation treatment is applied. Such analysis would consider pre- and post-treatment application periods (i.e., 0 to 10 years and 10 to 20 years). The analysis was made in two steps: one for a new conven- tional HMA pavement with a 10-year performance period and another for a 1-in. microsurfacing (assumed to be a 1-in. HMA overlay) of the existing HMA pavement. The condition of the pavement prior to application of the overlay would be taken as predicted performance of the pavement after 10 years. Except for an assumed 5% increase in base and subgrade moduli, all HMA layer properties, unbound base thick- nesses and properties, and subgrade layer properties were unchanged from the baseline design. Traffic volumes were adjusted to replicate the baseline design by using the same Data Category Data Element Analysis Parameters Design strategy—conventional flexible pavement Design life—20 years Performance Criteria and Reliability New flexible pavement performance indicators and reliability (assumed values) Condition Limit Reliability Initial IRI 60 in./mi — Terminal IRI 170 in./mi 90 Top-down cracking 2,000 ft/mi 90 Bottom-up cracking 25% 90 Thermal cracking 1,000 ft/mi 90 Total rut depth 0.75 in. 90 HMA rut depth 0.25 in. 90 Pavement Layers Layer types o HMA (CDOT grading SX) o Unbound base (CDOT Class 6) o Subgrade (A-2-6) Table 28. Baseline design inputs. Distress Distress Criteria Predicted Distress Achieved Reliability Terminal IRI, in./mi 170 138 99 Rut Depth – Total, in. 0.75 0.53 100 Rut Depth – HMA, in. 0.25 0.25 90 Bottom-Up Cracking, % 25 0.07 100 Top-Down Cracking, ft/mi 2,000 1,284 98 Transverse Thermal Cracking, ft/mi 1,000 27 100 Table 29. Baseline design predictions.

45 0 50 100 150 200 0 5 10 15 20 IR I (i n. /m i) Pavement Age (years) Threshold value Figure 16. Predicted IRI (90% reliability) for baseline design. 0.00 0.25 0.50 0.75 1.00 0 5 10 15 20 R ut D ep th (in .) Pavement Age (years) Threshold value Total rut depth HMA rut depth Threshold value (HMA) = 0.25 in. Figure 17. Predicted total rut depth (90% reliability) for baseline design. 0 300 600 900 1200 0 5 10 15 20 Th er m al C ra ck in g (ft /m i) Pavement Age (years) Threshold value Figure 18. Predicted transverse thermal cracking (90% reliability) for baseline design.

46 Pr ed ic te d Di st re ss o r I RI Age Baseline Design Preservation- Treated Design Extend Pavement Life Threshold for service life Figure 20. Illustration of the effect of preservation treatment application on pavement life. traffic characteristics for the first period (Years 0 through 10) and projected traffic volumes for the second period (Years 11 through 20). In this manner, the baseline and preservation-treated designs experience the same traffic loadings. Table 30 lists the inputs for the preservation-treated design. For these inputs, a 12-in.-thick pavement section is required to meet all performance criteria, consisting of 4-in. HMA grad- ing SX (PG 58-28) and 8-in. Class 6 aggregate base; a 1-in.-thick overlay (microsurfacing) will be applied after 10 years. The pre- dicted performance at 10 and 20 years is shown in Table 31. Plots for IRI, rut depth, thermal cracking, and total cracking (which includes reflective cracking and new bottom-up, top- down cracking) versus age are shown in Figures 21 through 24, respectively. Figure 21 shows that, although an increase in IRI is pre- dicted following the application of the treatment in Year 10, the predicted IRI remains below the threshold level over the 20-year design life. Figure 22 illustrates the predicted total rut depth at 90% reliability for the pavement (before and after preservation). The analysis assumes that the application of the 1-in. microsurfacing layer reduced the total rut depth (i.e., 0.50 in.) in Year 10 to zero. Figures 23 and 24 illustrate the predicted transverse thermal cracking and total cracking (at 90% reliability), respectively. The level of predicted cracking for the preservation-treated pavement is very low. 0 750 1500 2250 3000 0 25 50 75 100 0 5 10 15 20 To p- do w n Cr ac ki ng (f t/m i) B ot to m -u p Cr ac ki ng (% ) Pavement Age (years) Top-down Bottom-up Threshold value (top-down) Threshold value (bottom-up) Figure 19. Predicted fatigue cracking (90% reliability) for baseline design.

47 Distress Distress Criteria At 10 Years (prior to overlay/microsurfacing) At 20 Years (10 years after overlay/microsurfacing) Predicted Distress Achieved Reliability Predicted Distress Achieved Reliability Terminal IRI, in./mi 170 115 100 164 93 Total Rut Depth, in. 0.75 0.50 100 0.40 100 HMA Rut Depth, in. 0.25 0.18 100 0.06 100 Bottom-Up Cracking, % 25 0.16 100 1.45 100 Top-Down Cracking, ft/mi 2,000 1,635 94 1,394 97 Transverse Thermal Cracking, ft/mi 1,000 27 100 27 100 Table 31. Summary of distress prediction. Data Category Data Element Analysis Parameters New construction—conventional flexible pavement – Design life—10 years HMA overlay (microsurfacing) of conventional flexible pavement – Design life—10 years Performance Criteria and Reliability New flexible pavement performance indicators and reliability (assumed values) Condition Limit Reliability Initial IRI 60 in./mi — Terminal IRI 170 in./mi 90 Top-down cracking 2,000 ft/mi 90 Bottom-up cracking 25% 90 Thermal cracking 1,000 ft/mi 90 Total rut depth 0.75 in. 90 HMA rut depth 0.25 in. 90 HMA overlay (microsurfacing) of existing AC pavement performance indicators and reliability (assumed values) Condition Limit Reliability Initial IRI 90 in./mi — Terminal IRI 170 in./mi 90 Top-down cracking 2,000 ft/mi 90 Bottom-up cracking 25% 90 Thermal cracking 1,000 ft/mi 90 Total rut depth 0.75 in. 90 HMA rut depth 0.25 in. 90 Pavement Layers Layer types—new construction – HMA (CDOT grading SX) – Unbound base (CDOT Class 6) – Subgrade (A-2-6) Layer types—HMA overlay (microsurfacing) – 1-in. HMA overlay (microsurfacing) (CDOT grading SX) – New construction pavement section • • • • • • Table 30. Preservation-treated design inputs. Summary Analysis was conducted to estimate the effects of applying a microsurfacing (modeled as a 1-in. HMA overlay) in Year 10 of a 20-year design. The baseline design resulted in a pave- ment section consisting of 7 in. of HMA over 8 in. of aggre- gate base. The preservation-treated design was evaluated at 10 years (both prior to and after the application of micro- surfacing). The evaluation resulted in a pavement structure consisting of a 4-in. HMA layer on an 8-in aggregate base (with the 1-in. microsurfacing placed at Year 10). There are a number of issues that require further consideration: • The material properties and aging effects of the micro- surfacing were assumed to be the same as those of an HMA layer. To better evaluate the effects of microsurfacing treat- ment (or other treatment application), the treatment material properties, the potential changes to the existing layer(s), and aging effects need to be quantified. • Although the same cumulative number of trucks was assumed before and after the preservation application,

48 0 50 100 150 200 0 5 10 15 20 IR I ( in ./m i) Pavement Age (years) Threshold value = 170 in./mi Preservation- Treated Design Figure 21. Predicted IRI. 0.00 0.25 0.50 0.75 1.00 0 5 10 15 20 R ut D ep th (in .) Pavement Age (years) Threshold value = 0.75 in. Preservation- Treated Design Figure 22. Predicted rut depth. 0 300 600 900 1200 0 5 10 15 20 Th er m al C ra ck in g (ft /m i) Pavement Age (years) Threshold value = 170 in./mi Preservation- Treated Design Figure 23. Predicted transverse thermal cracking.

49 0 20 40 60 80 100 0 5 10 15 20 To ta l C ra ck in g (% ) Pavement Age (years) Preservation- Treated Design Figure 24. Predicted total cracking. conducting the analysis in two separate periods may not fully quantify the effects of repeated load applications and aging/climatic effects. • Increasing the resilient modulus of the unbound and sub- grade layers to account for the reduction in moisture infil- tration may not lead to appropriate consideration of the effect of a preservation treatment application. Although sealing of surface cracks and joints will minimize moisture infiltration, the effect of crack sealing on unbound and subgrade layer characteristics has not been established or considered in the EICM. To illustrate the potential effects of the microsurfacing treat- ment on the fatigue characteristics of the existing asphalt layer, analysis was conducted considering a softening or rejuvenat- ing effect of the treatment on the top portion of the exist- ing asphalt layer. Within the MEPDG, increasing the effective asphalt content by volume (Vbe) and lowering the percent air voids in the asphalt mixture (Va) will reduce the amount of predicted fatigue cracking but will increase rutting in the asphalt layer (ARA, Inc. 2004). For this analysis, 10% and 25% higher Vbe values (and cor- responding 10% and 25% lower Va values) were assumed for the existing asphalt layer. These changes resulted in very slight changes in the predicted distresses. Example 2: PCC Pavement Preservation Step 1: Identify Baseline Pavement Design and Preservation Treatments The specifics of the baseline pavement design are as follows: • Pavement type: JPC pavement • Design period: 30 years • Functional class—principal arterial • Traffic: – TTC, predominantly single-trailer trucks (TTC 1) – Two-way AADTT: 3,000 (assumed) – Number of lanes in the design direction: two – Percent trucks in design direction: 50 – Percent trucks in design lane: 95 – Vehicle class distribution and growth: default – Monthly adjustment: Default – Axles per truck: Default – Operational speed: 60 mi/hr – Axle distribution: Default – Axle configuration: Default – Lateral wander: Default – Wheelbase: Default • Closest weather station: Denver, CO Table 32 lists the recommended CDOT preservation treat- ments for JPC-surfaced pavements. Step 2: Identify Preservation Treatment Timing Diamond grinding treatment will be applied 20 years after original construction. Step 3: Identify Baseline and Preservation Treatment Material Properties The material properties and other parameters for the base- line pavement are based on the CDOT’s Standard Specifications for Road and Bridge Construction and Pavement Design Manual: • PCC – Unit weight: 150 lb/ft3 (default) – Poisson’s ratio: 0.20 (default)

50 – Thermal properties 77 Coefficient of thermal expansion: 5.5 in./in./°F × 10–6 (default) 77 Thermal conductivity: 1.25 BTU/hr-ft-°F (default) 77 Heat capacity: 0.28 BTU/lb-°F (default) – Mix 77 Cement type: Type I 77 Cementitious material content: 500 lbs/yd3 77 Water-to-cement ratio: 0.42 77 Aggregate type: Limestone 77 PCC zero-stress temperature: Calculated 77 Ultimate shrinkage: Calculated 77 Reversible shrinkage: 50% (default) 77 Time to develop 50% of ultimate shrinkage: 35 days (default) 77 Curing method: Curing compound – Modulus of rupture: 650 lb/in.2 – Surface shortwave absorptivity: 0.85 – Joint spacing: 15 ft – Sealant type: Liquid sealant – Doweled joints: No dowels – Widened slab: No – Tied shoulders: No – Erodibility index: Fairly erodible – PCC-base contact friction: Full friction with friction loss at 240 months – Permanent curl/warp effective temperature difference: -10°F • Unbound base: Class 6 – Aggregate type: Crushed stone – Poisson’s ratio: 0.40 – Coefficient of lateral earth pressure: 0.5 (default) – Resilient modulus: 38,721 lb/in.2 (CDOT average value) – Gradation (median of specification range): 100% pass- ing ¾ in., 47.5% passing No. 4, 40% passing No. 8, and 7.5% passing No. 200 – Liquid limit: 10 – Plasticity index: 2 • Subgrade: A-2-6 – Poisson’s ratio: 0.40 – Coefficient of lateral earth pressure: 0.50 (default) – Resilient modulus: 16,000 lb/in.2 (default) – Gradation: Default – Liquid limit: 15 – Plasticity index: 5 Step 4: Quantify Impact of Treatment Application on Pavement Thickness The diamond grinding application is assumed to reduce the thickness of the existing PCC by 0.25 in. Step 5: Identify Impact of Treatment Application on Existing Layer Material Properties Diamond grinding is assumed to have no effect on the existing pavement layer material properties. Step 6: Identify Immediate Impact of Treatment Application on Existing Condition It is assumed that diamond grinding will reduce faulting to zero and IRI to 90 in./mi. Treatment Type Distress Types Addressed Typical Thickness Comments Joint/Crack Resealing Cracking, joint seal damage Not applicable 1 to 4 years (typical) Diamond Grooving Macrotexture Not applicable — Diamond Grinding Faulting, roughness, macrotexture, pavement/tire noise, curling and warping, cross-slope 0.25 in. ADT, veh/day IRI, in./mi <3,000 90 3,000 to 10,000 76 >10,000 63 Partial-Depth Repair Localized surface distress Not applicable — Full-Depth Repair Severe spalling, joint/crack deterioration, full-depth cracks that divide a panel into two or more parts Not applicable — Cross-Stitching Poor load transfer at longitudinal joints Not applicable — Slab Stabilization Loss of support, faulting, corner breaks, settled slabs Not applicable — Dowel-Bar Retrofit Poor load transfer at transverse joints Not applicable — Table 32. Recommended preservation treatments for JPC pavements (CDOT 2013).

51 0 50 100 150 200 0 5 10 15 20 25 30 IR I (i n./ m i) Pavement Age (years) Threshold value Figure 25. Predicted IRI for baseline design. Data Category Data Element Analysis Parameters Design strategy—jointed plain concrete pavement Design life—30 years Performance Criteria and Reliability New concrete pavement performance indicators and reliability (assumed values) Condition Limit Reliability Initial IRI 60 in./mi — Terminal IRI 170 in./mi 90 JPC pavement transverse cracking 15% 90 Mean joint faulting 0.12 in. 90 Pavement Layers Layer types – PCC – Unbound base (CDOT Class 6) – Subgrade (A-2-6) • • • • Table 33. Baseline design inputs. Distress Distress Criteria Predicted Distress Achieved Reliability Terminal IRI, in./mi 170 149 97 Mean Joint Faulting, in. 0.12 0.12 90 Transverse Cracking, % slabs 15 4.4 100 Table 34. Summary of baseline design condition prediction. Step 7: Determine Dynamic Model The dynamic model will incorporate resetting the faulting to zero (see Figure 15) and the IRI to 90 in./mi upon diamond grinding application. Step 8: Develop a Baseline Design The material inputs listed in Table 33 were entered into the Pavement ME Design software program. The analysis determined that a 16-in.-thick pavement (10-in. PCC on 6-in. Class 6 aggregate base) will meet all of the performance criteria. The results of this analysis are listed in Table 34, and the predicted IRI, faulting, and panel crack predictions are shown in Figures 25 through 27, respectively (at a 90% reliability level). In this example, the level of faulting controls the recom- mended pavement design. IRI is predicted to reach a maximum value of 149 in./mi, the mean joint faulting is at the threshold

52 0 5 10 15 20 0 5 10 15 20 25 30 Tr an sv er se C ra ck in g (% sl ab s) Pavement Age (years) Threshold value Figure 27. Predicted slab cracking for baseline design. 0.00 0.05 0.10 0.15 0.20 0 5 10 15 20 25 30 M ea n Jo in t F au lti ng (in .) Pavement Age (years) Threshold value Figure 26. Predicted joint faulting for baseline design. value of 0.12 in., and transverse slab cracking is estimated to reach approximately 4% (all at 90% reliability). Step 9: Develop a Preservation-Treated Design No changes to the material inputs were assumed. The analy- sis considered pre- and post-preservation periods (i.e., 0 to 20 years and 20 to 30 years). For this example, the PCC thick- ness was reduced by 0.25 in. for the 20- to 30-year period, and the initial IRI was reduced to 90 in./mi. The analysis showed that a 15-in.-thick pavement (9-in. PCC on 6-in. Class 6 aggregate base) will meet all of the per- formance criteria if diamond ground after 20 years. The pre- dicted performance at 20 and 30 years is listed in Table 35, and the predicted IRI, faulting, and panel cracking are shown in Figures 28 through 30, respectively. Figure 28 shows that the predicted IRI remains below the threshold level over the 30-year design life, and Figures 29 and 30 show that mean joint faulting and transverse cracking stay below the respec- tive threshold levels before and after diamond grinding over the 30-year period. Summary Analysis was conducted to estimate the effects of apply- ing a diamond grinding treatment (modeled as a reduction in thickness and resetting IRI to 90 in./mi) in Year 20 of a 30-year design. The baseline design resulted in a pavement

53 Distress Distress Criteria At 20 Years (prior to grinding) At 30 Years (10 years after grinding) Predicted Distress Achieved Reliability Predicted Distress Achieved Reliability Terminal IRI, in./mi 170 140 98 163 93 Mean Joint Faulting, in. 0.12 0.11 94 0.09 99 Transverse Cracking, % slabs 15 4.49 100 4.39 100 Table 35. Summary of distress prediction. 0 50 100 150 200 0 5 10 15 20 25 30 IR I (i n./ m i) Pavement Age (years) Threshold value = 170 in./mi Preservation- Treated Design Figure 28. Predicted IRI. 0.00 0.05 0.10 0.15 0.20 0 5 10 15 20 25 30 M ea n Jo in t F au lti ng (in .) Pavement Age (years) Threshold value = 0.12 in. Preservation- Treated Design Figure 29. Predicted joint faulting.

54 0 5 10 15 20 0 5 10 15 20 25 30 Tr an sv er se C ra ck in g (% sl ab s) Pavement Age (years) Threshold value = 15 percent Preservation- Treated Design Figure 30. Predicted transverse cracking. section consisting of 10 in. of PCC over 6 in. of crushed stone base. The preservation-treated design was evaluated at 20 years (prior to grinding) and 10 years thereafter. The evaluation resulted in a pavement structure consisting of a 9-in. PCC layer on a 6-in. aggregate base, with the diamond grinding occurring at Year 20. Although the same cumulative number of trucks was assumed before and after the preser- vation application, conducting the analysis in two separate periods may not fully quantify the effects of repeated load applications.