Airfield and Highway Pavements 2017: Design, Construction, Evaluation, and Management of Pavements

The in-situ asphalt stiffness master curve is a critical input for flexible pavement design. The master curve can be established directly by performing dynamic modulus testing on field cores, predicted using the volumetrics of the mixture, binder grade and aggregate gradation established using field cores, or backcalculated using falling weight deflectometer testing. There can be significant differences between the master curves established using these three different methods. The effects of these differences on design were compared using distress predicted by the asphalt pavement design module in Pavement ME. It was found that the method used to determine the in-situ asphalt stiffness can have a significant effect on the predicted distress.

Traffic inputs for Mechanistic Empirical Pavement Design Guide (MEPDG) provides necessary information in terms of traffic load distributions, intensity and number of traffic load repetitions. The MEPDG requires more traffic data input than the AASHTO design guide, which makes switching to the new design guide a challenge to some states DOT’s. According to MEPDG, data inputs are a hierarchical as level 1, 2 and 3. Level-1 is site-specific data, level-2 is regional or statewide data, and level-3 is national wide data. This study compares the design output using site specific traffic input data to the design output using national default values. The analysis uses data collected and processed by Long Term Pavement Performance and measured distresses by Tennessee DOT. It was found out that the national wide inputs values overestimates the distresses. Also the site specific traffic input predicts more economical pavement sections than national wide although both input values estimate more economical designs than AASHTO-1993.

Rehabilitation of existing jointed plain concrete pavements (JPCP) by using unbonded jointed plain concrete overlay is increasingly being used all over United States. The design and performance of un-bonded overlays is affected by various factors including climate, traffic volume, concrete material properties, existing pavement condition and future design life. Modulus of elasticity/stiffness of concrete is an important material property that has significant effects on the performance of concrete pavements and many researchers have acknowledged it. But there has not been much research on effects of elastic modulus on performance of unbonded overlays. This study focuses on the impact of variability of elastic modulus of concrete on the mechanistic-empirical performance predictions of un-bonded JPCP overlays. Simulations were conducted using pavement ME design software with varying elastic modulus values and keeping all other design factors as constant. The results indicated that with decrease in elastic modulus values, the performance of un-bonded JPCP overlay is adversely affected with increased transverse cracking. The adverse effects of lower elastic modulus can be compensated by increasing the thickness of overlay concrete slab or by decreasing the transverse joint spacing and obtaining improved overlay performance.

Traditionally in the pavement design guides, the cumulative permanent deformation of multi-layered aggregate structures, such as highway pavements and airfield runways, is computed as a function of the number of load repetitions. This implies that the structural integrity of pavement foundations when subjected to one passage of extreme traffic loadings is overlooked. The passage of slow moving heavy traffic loads can potentially constitute a taxing scenario where the structural layers of the pavement undergo high levels of stress close to their failure limit. On the other hand, the anisotropic nature of granular materials imposes more complexity on this issue as the mechanism and probability of failure of such material can be substantially different from that of isotropic material. This paper addresses this problem by providing a mechanistic framework for the stability analysis of anisotropic unbound granular layers subjected to critical loading conditions. Based on the results of elastic analysis, the granular base will most probably remain stable if the material is assumed to be isotropic. However, the analysis of highly anisotropic systems with similar loading conditions resulted in unstable systems. This underscores the significance of departure from isotropic to anisotropic analysis for realistic simulation of the complex nature of the stress paths induced by traffic loadings.

The Pennsylvania Department of Transportation (PennDOT) has begun the process of implementing the Mechanistic Empirical Pavement Design Guide (MEPDG) and its accompanying software AASHTOWare Pavement ME Design into its routine pavement design practice through the characterization of traffic and material inputs, and verification/calibration of MEPDG transfer functions. This paper focuses on developing traffic inputs library for the use in AASHTOWare Pavement ME Design software. This paper identifies the truck traffic inputs used for evaluating the adequacy of a design strategy. PennDOT’s Roadway Management System (RMS) can generate most of the traffic inputs for a specific project. For roadway segments where project-specific traffic data are unavailable, the default values can be used for design. Wherever possible, these traffic default values were developed based on analysis of traffic data from PennDOT’s weigh-in-motion (WIM) and continuous automated vehicle classification (CAVC) traffic monitoring sites. However, for several truck traffic input parameters, supporting data were not available within PennDOT’s truck traffic database. For these parameters, the search for appropriate default values was conducted. The defaults based on the research-quality WIM data collected by LTPP SPS WIM TPF5 (004) study were found applicable for PennDOT designs. The traffic default values were included in the PennDOT Pavement ME Design data library. The PennDOT-specific traffic inputs will be used in the regional verification/calibration of transfer functions for Pennsylvania. They are required for predicting distresses in both flexible and rigid pavements.

The Utah Department of Transportation (UDOT) completed a study in 2009 to implement and adopt the American Association of State Highway and Transportation Officials (AASHTO) Mechanistic Empirical Pavement Design Guide (MEPDG). Prior to 2009, UDOT used a van mounted 3-point laser profiler to obtain the transverse profile of the pavement at regular intervals (typically a tenth of a mile). Maximum rut depth was then computed using the three points obtained. These measurements were found to be different than the LTPP definition through a direct correlation of rut depth measurements and an adjustment was made to the rutting of the UDOT sections so that they would match the LTPP rutting measurements. UDOT pavement sections rut depth measurement changed in 2009 to include closely spaced laser measured full transverse profile and calculation of maximum rut depth in each wheel path. This definition is thus very similar to that in LTPP. The change in rut depth measurements prompted UDOT to recalibrate MEPDG rutting model. The recalibration of the MEPDG rutting model was conducted using a nonlinear model optimization tool, SAS. A revised calibration coefficients were developed for rutting prediction model using Utah-specific design inputs and performance data. Overall, the recalibration process helped remove the bias (consistent over- or under-prediction), reduce the prediction error, and improve the accuracy of rutting model. The improvement in these models was notable; thus, indicating that the Utah-calibrated rutting model would provide much more accurate, reliable, and cost-effective designs than previous UDOT rutting prediction model.

Polycyclic aromatic hydrocarbons (PAHs) are known carcinogens, and previous research reported different concentrations of PAHs in the asphalt concrete (AC) pavements. The PAHs released from the asphalt binder and contaminate the soil and water over the time under different environmental and mechanical conditions. In this study, molecular dynamics (MD) simulation has performed to predict the release of Naphthalene PAH from asphalt binder component, Asphaltene Phenol in the presence of moisture and vehicle tire pressures. The release of PAH is predicted by computing energy losses under 0 to 5% moisture content and no vehicle tire pressure to 0.69 MPa tire pressure. The energy loss between PAH and Asphaltene Phenol molecules at 5% moisture and no tire pressure is 89% compare to 0% moisture. At 0.55 and 0.69 MPa tire pressures and at 5% moisture, the energy loss is 89 and 85% compared to 0% moisture, respectively. The results indicate that PAH will release less from AC pavements under high tire pressure but will release more under high moisture content.

The current Federal Aviation Administration (FAA) Advisory Circular does not address reflective cracking for asphalt overlaid concrete pavements. To implement this failure mode in flexible over rigid design procedure, a series of full-scale test pavements were constructed, instrumented, and tested at the FAA National Airport Pavement Test Facility. In this paper, laboratory experiments on field extracted hot mix asphalt (HMA) are conducted to study the characteristics of the mixtures for reflective cracking. Preliminary testing on the same mixtures was studied using the Texas overlay tester (OT), Indirect Tensile Strength, and Dynamic Modulus tests. The OT was customized to study the cooling effects on the initiation and propagation of reflection cracks on the above mentioned HMA cores. Findings from this research effort will be used to develop a rational testing procedure to simulate extreme cooling cycles through full-scale testing.

Extended finite element method (XFEM) models are prepared to observe the progression of an existing crack in the AC pavements under thermal fatigue load only. 2-D XFEM models are developed using the commercial software ABAQUS. The XFEM model consists of four layers, and it includes 263 mm asphalt concrete (AC), 150 mm base, 200 mm subbase, and subgrade layers. Creep tests were performed on the AC samples at −15, 0, and 18°C and the temperature-dependent viscoelastic material properties regarding Prony series were determined and used in the XFEM. The coefficient of thermal expansion (CTE) and contraction (CTC) of AC were also identified in the laboratory and the field. The pavement temperature data was collected using the temperature probes installed in the I-40 pavement section, and one-day temperature variation in the pavement is recorded, and that data is repeated as a thermal fatigue load in the XFEM model. Three XFEM models are developed considering one 50 mm existing crack but with three different cracks spacing such as two cracks are spaced within 1,220, 12,200, and 61,000 mm. The results show that the existing crack propagation stops at approximately 5 thermal cycles for the cracks that have 1,220 mm crack spacing, 6 thermal cycles for 12,200 mm crack spacing and 8 thermal cycles for 61,000 mm crack spacing. The crack length propagated from the existing 50 mm length to 81% for 1,200 mm crack spacing, 106% for 12,200 mm crack spacing and 118% for 61,000 mm crack spacing.

The finite elements method (FEM) is a numerical method that thanks to its versatility and adaptability, allows for the modelling and simulation of the entire set of hypothesis related to the complex morphology, load, boundary and support constraints of road and highway infrastructures. This document presents the StratCalc finite elements analysis model, a model that was specifically developed in order to be applied for the analysis of pavement structures mechanical behaviour and correspondent performance prediction. In this paper, StratCalc is used for load format comparison (between the classic circular loaded area and equivalent oval, rectangular and square load formats) of two test pavement structures. Through interpretation of the numeric and graphic output of the StratCalc analysis model, it is possible to establish that regardless of the considered load format, the calculated displacements, strains and stresses tend to converge. It is also clearly shown by the evolution of the vertical pressure bulbs and strain isolines that this convergence is gradually achieved after load degradation (in both the horizontal and vertical directions). Depending on each pavement structure characteristics and properties, the varying gradual load degradation process associated with each load format might represent significant variations on the maximum strains and stresses applied at the pavements critical points, and consequently, to the structures expected lifespan.

The objective of this study was to perform a statistical analysis on pavement response data collected from accelerated loading testing at the Virginia Tech Transportation Institute. A heavy vehicle simulator device (model Mark VI from Dynatest) was used to load a test section of asphalt pavement featuring a cold central plant recycled base layer. The test section was subjected to different loading conditions to explore the effect of speed, wheel load, and tire inflation pressure on the pavement responses. Pavement response data was collected through a data acquisition system with an instrument set of 3 pressure cells and 3 strain gauges. A three level full factorial experiment was conducted using speed, wheel load, and tire inflation pressure, which resulted in a total of 27 different loading configurations. Analysis of variance and correlation analysis are conducted on the instrument data to better understand the effects speed, wheel load, and tire inflation pressure on pavement responses. It was found that the instrumentation recorded signals that were repeatable under varying loading conditions. Also, as expected, the pressure and strain responses were highly correlated to speed and wheel load.

The construction quality of a lot is often measured by collecting and testing a few samples. The statistics of an acceptance quality characteristic (AQC) of the lot determined based on the samples is used to make an inference on the overall construction quality. The most commonly used quality measure to quantify the quality is percent within limits (PWL). For performance-related construction specification, the PWL value of a lot is related to its expected future performance. While a contractor is paid an incentive if the measured quality is better than acceptable quality limits (AQL), the contractor may be penalized if the measured quality is less than AQL. Therefore, the accuracy of PWL estimation is important to minimize over/under payment to the contractor for the produced quality. In this paper, an effort has been made to determine the effect of sampling plans and sizes on the PWL estimation of a lot. Air void contents after construction were used as the AQC with two sided specification limits. Different sampling plans include: (1) random sampling with replacement, (2) stratified sampling, and (3) random sampling without replacement. For each sampling plan, samples of sizes 3, 5, 10, and 20 were used to evaluate the combination (i.e., sampling plan and sample size) for estimating PWL. The estimated PWL values were compared with the true PWL of the lot. The results show that stratified sampling plan represents the true construction quality of a lot better for lower sample sizes.

Pavement preservation can be a very effective tool to extend functional and structural integrity of existing highways before they need major rehabilitation or replacement. There are no general guidelines available for developing performance-related specifications (PRS) for pavement preservation treatments that correlate key engineering properties to treatment quality and long-term performance. This paper documents the process to facilitate development of PRS for pavement preservation treatments that provide a direct relationship of key construction characteristics to the performance. The development process for preservation treatments has unique differences when compared with that of rehabilitation and new construction of pavement. The differences include; non-structural nature of preservation treatments, functional performance measures, and construction materials and processes. The process developed in the paper is demonstrated with the help of an example for micro-surfacing treatment. Micro-surfacing treatment can improve surface friction and address minor surface irregularities. Typically, the treatment is employed to address surface friction problem on an existing asphalt surface. Mean profile depth (MPD) was selected as an acceptance quality characteristic (AQC) because it is an essential component of macro-texture and surface friction (performance measure) and can be measured after construction by using high-speed laser measurement devices. Simulated pavement data were used to demonstrate a relationship between the AQC (before and after the treatment) and expected performance (life extension). The relationship was used to develop pay factors for pay adjustments (i.e., incentives and disincentives). The example also demonstrates the use of quality measure limits and their impact on the contractor’s and agency’s risks.

Road conditions performance modeling is required in order to predict the future conditions and provide information that can be applied to transportation planning, decision making processes and identification of future maintenance interventions. As extension of knowledge in existing gravel road condition models, improved artificial intelligent gravel road performance models which best capture the effects of gravel loss condition influencing factors were developed using feed forward neural network (FFNN) hybrid with a district GIS-based map using linear referencing approach to display gravel loss conditions as a threshold to trigger optimal maintenance interventions. The developed FFNN gravel loss condition (GVL) prediction model yielded R2 = 0.95 > 0.9 benchmark based on minimum MSE = 0.055 < 0.1. Threshold value = 3 (fair condition) was specified on the GIS map for triggering maintenance interventions when gravel road subgrade exposure due to gravel loss is between 10 and 25% as condition monitoring innovative tools.

Due to the complexity and diversity of pavement surfaces, pavement cracking detection is a challenging task even for human operators. The automation of pavement cracking detection generally requires robust algorithms with high level of intelligence. From such a perspective, Deep Learning, a promising branch of Artificial Intelligence, can serve as an advanced approach to high level of intelligence by learning from huge amount of historical data and enhancing the capability of behaving correctly under unforeseen and complex environments. This paper proposes a convolutional neural network (CNN) for recognizing cracks on asphalt surfaces at subdivided image cells. The proposed CNN consists of 3 convolution layers and 2 fully-connected layers with 1,246,240 parameters in total. Feeding the proposed CNN with 5,000 manually-processed example images recursively, the accuracy of the proposed CNN is progressively improved, and the generalization is also enhanced. It is demonstrated in the paper that the trained CNN can achieve high accuracies 96.32 and 94.29% on training data and testing data respectively.

In order to better understand the evolution of the road-surface macro-texture, kneading test was conducted by using a kneading machine. The kneading machine was invented by improving a national standard rutting machine. Ten specimens were studied, which have different values in various factors (kneading temperature, aggregate gradation and aggregate type). Macrotexture measurements were carried out at different kneading times (0, 2, 4, 6, and 8 h). The macro-texture profiles were measured by means of the Laser-measuring Technique, showing on an area of 10 × 10 cm. Two indicators, the Peak Angle and the Kneading Loss, were adopted to describe and quantify the evolution of the macro-texture. The results showed that aggregate gradation, kneading time and kneading temperature can significantly impact the skid resistance performance of asphalt pavements; Kneading will affect the distribution of Peak Angles and increase the Kneading Loss.

Roadway agencies in Canada are at various levels of maturity in the implementation of transportation asset management programs and systems. While provincial highway agencies and the larger cities generally are more advanced because they have more resources than smaller municipalities, strong leadership and champions within the agencies are moving forward to implement comprehensive systems to assist in the stewardship of our aging infrastructure. There is a strong movement in Canada for agencies to integrate investment planning and programming and move away from traditional silo-based infrastructure management systems into one comprehensive system. Agencies are moving toward systems that track asset performance from construction to retirement and use life cycle cost to make whole life maintenance and rehabilitation investment decisions. These systems report on achievements in maintaining and improving asset condition as measured by key performance measures. There are significant differences across Canada in terms of the scope of agency asset management systems. Some agencies have a strong desire and leadership for asset management while others are hesitant to change. Others have constituted asset management groups who are responsible for all of the functions of the asset management system. The level of competence of asset management support varies across agencies. Many have reported difficulty in maintaining consistency of staff in asset management positions, mainly due to attrition or due to lateral moves within the agency. The method and degree of communication of asset management activities varies significantly across Canada. Most agencies have developed some form of annual “report card” to communicate progress on the effective management of assets. For municipal agencies, the report cards are geared toward the public and are used to assist in establishing funding support for long-term sustainable asset management. This paper outlines the findings, best practices and lessons learned from a review of the detailed transportation asset management practices of 25 provincial, municipal, trade organization and private sector road and light rail agencies.

In the State of Connecticut, approximately 35% state maintained roads were built prior to 1950, while 44% were built between 1950 and 1980. These roads are typically maintained through multiple patching and resurfacing projects resulting in a large network of composite pavement. Composite pavement failure observed in Connecticut is generally joint failure in the underlying rigid slabs. The DOT (CTDOT) piloted a non-destructive testing approach to identify concrete repairs prior to removal of the asphalt surface to help develop better estimates, leading to fewer costly change orders. The pilot approach included a comprehensive ground penetrating radar (GPR) program, visual review of surface cracking, and limited falling weight deflectometer (FWD) and boring programs. GPR and FWD joint and concrete condition results showed good correlation CTDOT will finalize the pilot by assessing the PCC material and joint condition at the time of asphalt removal during construction, compared to the NDT results.

India has a large network of roads of length over 5 million kilometres, consisting of different categories of roads. The maintenance management of such large road network with limited resources is a demanding task in emerging economics like India. The road users expect the performance of these road infrastructures to be above the desired level during the service life. The road authorities expect this performance to be achieved within the limited resources. This contradicting inevitability makes the problem quite interesting and challenging. In the present study, a framework for the maintenance management of road network through multi-criteria approach is proposed by considering the (performance) deterioration level as an objective function which is to be minimized. For this purpose, two distress criteria are considered, one reflecting functional condition of pavement quantified in terms of roughness and the other reflecting structural condition of pavement quantified in terms of rebound deflection. Development of a decision support system minimizing the roughness and the deflection by selection of appropriate maintenance strategies at the optimal time for different categories of roads carrying different volumes of traffic, within available resources (budget) is a challenging task. A multi-objective optimization model is developed using mixed integer linear programming (MILP). The ε-constraint method is adopted by considering one of the distresses as constraint to generate non-dominated solutions. A knee point is also identified as the best available solution among the Pareto-optimal solutions. The framework can be adopted for preparation of an annual budget or a block year budget within the maintenance horizon.

Proper functionality of pavement structures can be ensured with adequate quality assurance and quality control during construction. By monitoring pavement health throughout its design life, authorities can reduce structural damages, and achieve subsequent reduction in maintenance cost and time. Geophysical investigation is growing significantly as a tool of QA/QC of pavement structures. The increased use can be attributed to (1) decreased time consumption, (2) reduced expense, (3) lower variability of test results compared to laboratory tests, and (4) the non-reduction of structural integrity and serviceability of pavement compared to destructive testing. The current research work evaluated the applicability of the slab impulse response (SIR) method in pavement health monitoring by conducting in-situ test results at four different locations of a construction project in Dallas, Texas. The dynamic cone penetrometer (DCP) and Geogauge were utilized alongside SIR to determine the resilient and elastic modulus of the pavement layers. Field investigation results indicated that SIR can be useful in identifying damages associated with low stiffness, such as delamination, honeycombing, cracking, and voids below the slabs-on-grade. Comparisons presented in this paper demonstrated the effectiveness of SIR as an alternative tool to conventional methods of pavement health monitoring.

MIT Scan technology is a non-destructive method to measure the alignment of dowel bars placed in transverse joints of jointed plain concrete pavements (JPCP). Misalignment of dowel bars can potentially lead to reduced load transfer efficiency (LTE), cracking, spalling and faulting. Several agencies across the U.S. have adopted specifications for dowel bar misalignment, many using the joint score methodology. However, this methodology has never been validated using field studies. National Cooperative Highway Research Program (NCHRP) Report 637 provides a methodology to determine an effective dowel diameter; however, this was based on laboratory testing and limited field studies. This paper presents results of a Federal Highway Administration (FHWA) project to collect MIT Scan data on long term pavement performance (LTPP) sections. The analysis did not indicate any definitive relationship between joint score and cracking/spalling within the analysis range for most States. This is not to say that severely misaligned dowel bars do not affect pavement performance, particularly localized distresses. Analysis of effective dowel diameter as a measure of dowel misalignment for use with Pavement ME suggests that using effective dowel diameter is a less biased estimator of long-term LTE as modeled using Pavement ME than using actual dowel diameter, thus suggesting a relationship between dowel misalignment and long-term LTE. The remaining bias and scatter suggests that models (effective dowel diameter model and LTE model in Pavement ME) can be improved using the data collected as part of this study.

The STAC (French Civil Aviation Technical Center) elaborated a methodology for flexible pavements assessment using HWD. It is described in a technical guidance, released in 2014. It emphasizes the importance of complementary operations to be performed during the survey, especially temperature measurements in the asphalt materials. Backcalculated parameters correspond actually to the material properties within the test conditions, i.e. for the in situ temperature. Default temperature corrections are proposed in the guidance, which correspond for each material type to standard values. A 1-year test survey was conducted in 2014–2015 with the goal of validating these corrections or improving them. It consisted in repeated HWD tests, performed on 10 identified test points from the STAC’s instrumented test facility, which includes several embedded temperature sensors. Surveys were conducted every two weeks, during a full year, providing a significant deflections vs temperature data base, for a full temperature range with regards to French climate. This paper first presents the qualitative analysis performed from the raw data, especially about the relative influence of temperature on center and outer deflections. Then the numerical study conducted is presented. It included for each survey dynamical backcalculations using deflections from the corresponding representative test point. Finally the backcalculated asphalt materials moduli from all survey corrected with regards to temperature are compared. The experiment allowed demonstrating that the standard corrections provide fairly consistent results, but that the model can be refined.

This study is done to predict the International Roughness Index (IRI) for flexible pavements using climate and traffic data by employing artificial neural network (ANN) modeling. The climate and traffic data are collected from the long-term pavement performance (LTPP) database. The ANN model is trained using 50% of climate, traffic, and IRI data and then rest 50% data is used to validate the model by comparing ANN predicted IRI and measured IRI for flexible pavement under a climatic zone. The trained model and the validated model are compared by calculating root mean square error (RMSE) of ANN predicted IRI and measured IRI. A flexible pavement located at the wet-freeze climatic zone, employing 7-7-1 ANN structure and using Pure Linear transfer function, the RMSE generated is 0.055. A better prediction ANN model is generated using 7-9-9-1 architecture using a non-linear transfer function and the RMSE further improved to 0.012.

State Highway Agencies track the roughness of their pavement network using the International Roughness Index (IRI). In addition, many state highway agencies use the IRI to judge the smoothness of new construction for construction acceptance. Data collected by an inertial profiler is used to compute the IRI. An inertial profiler must be certified to ensure that it obtains repeatable and accurate IRI values. Some highway agencies evaluate the repeatability and accuracy of profilers by using criteria established by the agency, while others use the procedures described in AASHTO standard R 56. This paper will present common procedures that are used to evaluate the repeatability and accuracy of inertial profilers, including the method specified in AASHTO R 56. Common errors that can cause a device to not meet the AASHTO R 56 criterion are also described. The height sensor types used in inertial profilers are single-spot lasers and line lasers, with single-spot lasers not being able to collect accurate data on longitudinally textured surfaces. The paper addresses how the data collected by a single spot-laser and a line laser compares with each other on various surface types.