Mechanistic-Empirical Pavement Design Guide Research Articles

Nonlinear Interpretation of Resilient Modulus Parameters Considering Incremental Static and Dynamic Loading

The resilient modulus ( Mr) is typically interpreted as the average of the last five secant slopes of the cyclic stress-axial resilient strain curve from repeated loading triaxial tests. It is not uncommon to then fit mathematical models to the secant Mr to derive model parameters ( Ki, more commonly known as K1, K2, and K3) that are then used in pavement analysis. Some engineers also backcalculate Ki from a falling weight deflectometer test. Many researchers use an incremental loading procedure to analyze pavements. When doing so, it is important to consider the nonlinear load-deformation behavior, which is considered only coarsely in the secant slope approach. Some incremental loading procedures assume the geomaterial to be nonlinear elastic while a dynamic finite element analysis typically assumes the geomaterials to be nonlinear and viscoelastic, that is, having springs and dashpots. With the latter, additional damping parameters to represent the viscoelastic effects are required. This paper compares Ki parameters estimated using linear regression on a log transformation of the Mechanistic–Empirical Pavement Design Guide (MEPDG) Mr model, which is a non-viscoelastic secant Mr approach, with those obtained using nonlinear regression of the time dependent deformations when interpreting the Mr test using incremental loading of a: 1) nonlinear elastic geomaterial; and 2) nonlinear viscoelastic geomaterial with inertial mass. The results show that the estimated Ki parameters governing the geomaterial nonlinearity are substantially affected by the estimation method and that the second alternative approach can model hysteresis well.

Integrating Mechanistic-Empirical Pavement Analysis in the Life Cycle Assessment Use Phase and Monetization of Environmental Impacts to Promote Low Carbon Transportation Materials

This study attempted to integrate the Mechanistic-Empirical Pavement Design Guide (MEPDG) analysis into the life cycle assessment (LCA) framework to explore the potential of terminal blend rubberized bitumen (TBRB) technology as a sustainable alternative to traditional paving materials in the United Arab Emirates (UAE). In addition, the concept of environmental impact monetization is introduced for quantifying shadow prices to be used in cost–benefit analyses and encourage the adoption of sustainable technologies in pavement construction. The findings demonstrate that hot mixed asphalt (HMA) with TBRB outperforms asphalt mixtures with neat bitumen (NB) and polymer-modified bitumen (PMB) containing 7% linear styrene–butadiene–styrene, with respect to durability and environmental impacts. The cradle-to-grave analysis, which includes maintenance and rehabilitation cycles based on the MEPDG results, alters the hierarchy among scenarios compared to conventional cradle-to-gate analyses, highlighting the importance of the use phase in LCA. Overall, the asphalt mixture with TBRB exhibits 83% and 63% less global warming potential than mixtures with NB and PMB, respectively. The monetization process through the distance-to-target approach returned the shadow price of US$157 per metric ton of CO2eq in the UAE. Based on this value, the HMA with TBRB, if selected over those with NB and PMB, offers environmental savings of US$47,186/km-lane and US$15,852/km-lane for the case study investigated.

Impacts of climate change on environmental and economic sustainability of flexible pavements across China

Pavements are susceptible to accelerated deterioration due to changing climate conditions, leading to increased maintenance and excess fuel consumption through pavement-vehicle interaction. China's diverse climates raise concerns about the environmental and economic sustainability of flexible pavements amid climate change and the effectiveness of preventative maintenance strategies. This study examines climate change's potential impacts on long-term pavement performance, greenhouse gas (GHG) emissions, and costs, employing the Mechanistic-Empirical Pavement Design Guide method. The calibrated World Bank's Highway Development and Management Model 4 assesses the impacts of surface characteristics on vehicle fuel consumption. Life cycle assessment and life cycle cost analysis quantify GHG emissions and costs. Results reveal significant pavement deterioration in Southeast and Central China. Preventative maintenance strategies reduce fuel consumption, with GHG emissions and cost savings from smoother driving conditions outweighing those from maintenance. These insights stress the importance of proactive maintenance strategies for mitigating climate-induced deterioration and enhancing sustainability.

Bus Lane Design Based on Actual Traffic Loads and Climate Conditions

Bus lanes play a crucial role in urban areas as their primary objective is to increase public transport efficiency and help traffic and public transit systems flow more smoothly. This study starts with traffic and climate monitoring to verify asphalt bus lanes in Rome, Italy, according to the Italian Pavement Design Catalogue published in 1995. KENLAYER software calculated the stress-strain conditions under real traffic loads (i.e., hourly passages of urban buses, considering their axle load and seat occupancy rate), typical subgrade bearing capacity (i.e., resilient modulus equal to 90 MPa), current climate conditions, and road material properties. Then, the Mechanistic-Empirical Pavement Design Guide (MEPDG) was used to verify the response of the pavement structure. The fatigue verification of bound materials resulted in damage values much lower than 1 at the end of the 20-year service life (i.e., 0.12 with the Asphalt Institute and 0.31 with the Marchionna law, respectively) and highlights that the Italian catalogue’s sheets are overdesigned. On the other hand, the rutting verification according to MEPDG is not satisfied after an 11-year service life (i.e., the total rutting is equal to 1.50 cm), forcing frequent and expensive maintenance of wearing and binder courses. Therefore, the results confirm the validity of the Italian catalogue for fatigue service life and suggest the need for high-performance asphalt to prevent early rutting due to bus traffic increasing by load and frequency in previous decades.

Quantifying the Effects of Material Input Levels on Jointed Plain Concrete Pavement (JPCP) Performance and Slab Thickness

The mechanistic-empirical pavement design guide (MEPDG) is a commonly accepted design principles guide that aids in jointed plain concrete pavement (JPCP) design and performance analysis. The MEPDG uses three different design parameter input levels, referred to as level one, level two, and level three, providing increasing confidence in the analysis at the lower numbered levels, which use more locally relevant (level two) or project-specific (level one) data. The state-of-the-art pavement ME software (version 2.6.2) uses MEPDG design principles to predict pavement performance. The three performance indicators for JPCP systems (international roughness index (IRI), joint faulting, and transverse cracking) experience significant changes when simulating under a different input level. The IRI and faulting indicator changed by 78 percent when using inputs varying from level one to level three, with the cracking indicator change being more severe at 87 percent. To accommodate the change in performance indicator values between input level one and input level three, increasing the concrete slab thickness is necessary to achieve comparable pavement performance. An increase in the Portland cement concrete (PCC) layer from one inch to two inches is required when input level three simulations are performed, demonstrating the economic and sustainability benefits of using project-specific level one inputs. Understanding the impact of simulation input levels will help to meet design and sustainability goals and improve the lifecycle performance of JPCP systems.

Using Repeated Light-Weight Deflectometer Test Data to Predict Flexible Pavement Responses Based on the Mechanistic–Empirical Design Method

This study investigated the potential of lightweight deflectometer (LWD) data in predicting layer moduli and response measurements within the Mechanistic–Empirical Pavement Design Guide. To achieve this goal, field repeated LWD tests and laboratory repeated load triaxial tests were carried out on granular base material compacted at 3% and 6% water content, sandy subgrade soil compacted at 3%, 4% and 9% water content and silty sand subgrade soil compacted at 8% and 10% water content. The results revealed that substituting traditional repeated load triaxial (RLT) data with LWD data for predicting these parameters was notably effective for cohesionless materials, especially for unbound granular materials (UGMs) compacted at optimum water content. The accuracy and reliability of predictions were remarkably high, showcasing the potential of LWD to enhance efficiency and precision in pavement design within this context. Conversely, for cohesive road materials, the study emphasized the importance of considering specific material properties and water content when integrating LWD into the Mechanistic–Empirical Pavement Design Guide. The distinctive characteristics and behaviors of cohesive materials necessitate a nuanced approach. This understanding is critical to ensuring the accuracy and reliability of pavement design and assessment across diverse conditions. In summary, the study presents a promising avenue for utilizing LWD data in cohesionless road materials, offering potential cost and time-saving advantages. Additionally, it underscores the necessity of tailored approaches when considering material properties and moisture content for cohesive materials, thereby advancing the field of pavement engineering by providing insights for improved practices and adaptable frameworks.

Laboratory Evaluation of Marginal and Industrial Waste Material in Geocell-Reinforced Pavements under Cyclic Loading

Abstract The availability of quality materials for the construction of pavements has been a problem in some regions. This scarcity enforced geosynthetic materials as one of the quests for su1stainability in the pavement industry. This study attempted to stabilize marginal (tunnel muck, reclaimed asphalt pavement material) and industrial waste (zinc slag) materials by employing high density polyethylene geocell to appraise the effectiveness of geosynthetic reinforcement in enhancing the characteristics of pavement materials and mitigating their rutting. The cyclic plate load tests are performed on marginal and industrial waste material with geocell-reinforced and unreinforced base layers over black cotton soil as the subgrade. The tests were conducted following the trapezoidal loading pattern with a 0.77-Hz frequency. The geocell-reinforced pavement performance was evaluated for resilient deformation, accumulated permanent deformation, rut depth reduction, traffic benefit ratio, and base layer thickness reduction. The mechanistic empirical pavement design guide (MEPDG) permanent deformation performance (PDP) model framework was used to predict the accumulated permanent deformation in unbound granular layers and to differentiate the rutting performance in geocell-reinforced and unreinforced sections. This study determined the material constants of the MEPDG PDP model for marginal and industrial waste materials used in the pavement section for quantifying the reduction in base layer thickness.

Carbon emissions quantification and different models comparison throughout the life cycle of asphalt pavements

Establishing a comprehensive and systematic method for quantifying carbon emissions (CEs) of asphalt pavement is challenging due to the multifaceted and diverse nature of the involved processes. Utilizing a life cycle assessment framework, this study introduces a practical model system for capturing CEs across the entire lifespan of asphalt pavement. A pivotal aspect of our approach involves calculating activity levels. Material quantities are ascertained based on mixture proportions and specific pavement information. We employ Ridge Regression and the quota method to approximate the energy consumption of construction machinery. A comparative analysis is performed between two CE computation models that use energy consumption and operational time as activity levels, respectively. Additional vehicle fuel consumption is translated based on shifts in the International Roughness Index, as predicted by the Mechanistic-Empirical Pavement Design Guide. CEs from traffic disruptions are calculated collaboratively through VISSIM simulations and MOVES emission modeling. Our findings reveal that material production, mixture preparation, and maintenance duration are the principal factors contributing to CEs throughout the asphalt pavement lifecycle. Notably, emission estimates for machinery in the construction phase can vary by as much as 8.3 times across different models. The study suggests that incorporating the use of recyclable materials and designing more durable pavement structures can offer effective avenues for CE mitigation.

Resilient response and strength of highly expansive clay subgrade stabilized with recycled concrete aggregate and granulated blast furnace slag

In urban areas, due to urban development and lack of space, roads may be built on the expansive clay subgrade. Expansive clay soils increase in volume when in contact with water and shrink by losing water. Recycled concrete aggregates (RCA) are one of the components of construction and demolition (C&D) wastes that can be used instead of natural resources in road construction projects. In this research, a combined method of mechanical and chemical stabilization was used to evaluate the use of RCA as a mechanical stabilizer and granulated blast furnace slag (GBS) as a chemical stabilizer to improve the strength and resilience characteristics of expansive clay subgrade. Laboratory results showed that adding RCA wastes to expansive clay subgrade reduced atterberg limits (LL, PL, and PI) and optimum water content (OWC) and increased the maximum dry density (MDD), California bearing ratio (CBR), and uniaxial compression strength (UCS). Expansive clay subgrade with 20 % RCA was treated with GBS chemical stabilizer in order to achieve a minimum 7 days target strength of 1.7 MPa for subbase application. A series of repeated load triaxial (RLT) tests were performed according to the AASHTO T307 procedure and resilient modulus (Mr) values of the stabilized clay subgrade specimen were evaluated at different curing times under cyclic loading. Also, the proposed Mechanistic-Empirical pavement design Guide (MEPDG) model was used to predict Mr values in different confining and deviator stresses which showed a good correlation with the measured Mr values. Finally, the improvement of the clay subgrade stabilized with RCA and GBS was also supported by SEM analysis. The results obtained from this research showed that the use of RCA, in addition to reducing disposed waste and environmental pollution, can improve the physical and mechanical characteristics of expansive clay Subgrade layers.

Reliability analysis of flexible pavements based on the quantile-value method

ABSTRACT This paper presents a simple yet efficient approach for the reliability-based design of flexible pavements, based on the concept of quantile-values. The reliability-based approaches currently adopted in the pavement design guides are simplified techniques, based on an overall reliability factor. On the other hand, simulation-based probabilistic techniques, widely discussed in literature, can be computationally expensive and tedious to implement. The ‘Simplified Effective Random Dimension Quantile Value Method’ (Simplified ERD-QVM), adopted in this study, presents an elegant technique to bridge the gap between the precision of simulation-based probabilistic techniques and the efficiency of design guide approaches. The accuracy of this technique is verified through Monte Carlo Simulations for different pavement design scenarios. The main advantages of the method are (i) the significant savings in computational times for reliability-based design of flexible pavements without the use of surrogate models (ii) the use of simple techniques already familiar to practitioners who adopt the Mechanistic-Empirical Pavement Design Guide and (iii) it is a comprehensive reliability analysis approach that addresses the drawbacks of the currently used approaches, while capturing the effect of both design parameter and model uncertainty. The methodology for integrating the Simplified ERD-QVM with the existing flexible pavement design codes is also presented.

Utilizing expansive soil treated with phosphogypsum and lime in pavement construction

This study investigated the use of phosphogypsum (PG) in the long term as a construction material, especially in pavement structures. This study aimed to investigate the effect of increasing PG percentages on the geotechnical parameters of Expansive Soil. Extensive laboratory tests were conducted on both natural and PG-stabilized soil mixtures to examine their compaction properties, consistency limitations, swelling and linear shrinkage, pH values, and unconfined compressive strength (UCS), California bearing ratio (CBR), scanning electron microscopy (SEM), and X-ray Diffraction (XRD). The results showed that the PG addition would significantly reduce soil plasticity, linear shrinkage, and swell potential. Despite these improvements, the study concluded that treating the soil using PG only did not affect its physical properties sufficiently to be considered as a pavement subbase. As an alternative, lime was supplemented to the soil-PG mixture and noticed that a blend of 30% PG and 4% lime significantly improved soil strength, thereby rendering it appropriate for utilization as a pavement subbase layer. The Mechanistic-Empirical Pavement Design Guide (MEPDG) was utilized to assess the performance of pavement sections formed with PG-treated and untreated soils. As a result, this study contributes to the development of an environmentally sustainable PG waste management system while also demonstrating its potential uses in roadway construction.

Effect of the Field-Stress State on the Subgrade Resilient Modulus for Pavement Rutting and IRI

The new Mechanistic-Empirical Pavement Design Guide (MEPDG) uses the subgrade resilient modulus (MR) as the key input parameter to represent the subgrade soil behavior for pavement design. The resilient modulus increases with an increase in confining pressure, whereas, for an increase in deviatoric stress, it increases for granular soils and decreases for fine-grained soils. The value of MR is highly stress dependent, with the stress state (i.e., bulk stress) a function of the position of the materials in the pavement structure and applied traffic loading. Applying excessive vertical stress at the top of the subgrade without knowing the appropriate stress state can result in permanent deformation. In situ stress must be calculated so the correct resilient modulus can be determined. To facilitate the implementation of MEPDG, this study develops a methodology to select the appropriate subgrade resilient modulus for predicting rutting and IRI. A comprehensive research methodology was undertaken to study the effect of in situ or undisturbed subgrade MR on pavement performance using the MEPDG. Results show that MR obtained from in situ stress is approximately 1.4 times higher than the MR estimate from NCHRP-285. Thus, the in situ stress significantly affects the calculation of subgrade MR and, subsequently, the use of MR in the predicted rutting, with IRI using the AASHTOWare pavement mechanistic-empirical design. Results also show that the pavement sections were classified as in “Good” and “Fair” conditions for rutting and IRI, respectively, considering in situ MR.

Unbonded Concrete Overlay Critical Stress Prediction for Conventional and Short Width Lanes

Accurate critical stress prediction is a key element for a mechanistic–empirical design of pavement structures. For unbonded concrete overlays (UBOL), the critical stress prediction procedure used by the current Mechanistic–Empirical Pavement Design Guide (MEPDG) has several shortcomings. The current MEPDG UBOL model only predicts top or bottom-critical longitudinal stresses, so transverse fatigue cracking is considered whereas longitudinal cracking, which has been observed in several in-service overlays, is ignored. The MEPDG critical stress prediction models assume that the UBOL and the existing pavement have the same deflection profiles, ignoring the separation between the overlay and the existing pavement. Various interlayer properties influencing critical overlay stress are also overlooked. In addition, the model only considers conventional lane width UBOLs and is unable to analyze short slab overlays. The latter has been gaining popularity as an attractive geometry for UBOLs. This paper documents the development of an improved procedure for computing UBOL critical stresses for conventional and short width slabs geometry. The procedure is able to predict critical transverse and longitudinal stresses at various locations taking into account interlayer stiffness and the presence of voids in the interlayer. Rapid solutions using neural networks were developed to enable computationally efficient UBOL critical stress prediction.

Fatigue Cracking Characteristics of Asphalt Pavement Structure under Aging and Moisture Damage

Structural characteristics influence assessment of fatigue cracking behavior. In the assessment of asphalt pavements, the asphalt structure and practical conditions must be considered. This study analyzes changes in the elastic modulus of the pavement of different asphalt mixtures amid aging and moisture damage through fatigue cracking tests. A model to predict the tensile strain at the bottom of the pavement layer is developed through a structural analysis based on the material properties. The results are comparatively analyzed using the Mechanistic-Empirical Pavement Design Guide to predict the fatigue crack life. The test results indicate that moisture damage significantly influences the material properties of asphalt pavement and can accelerate pavement damage as the asphalt ages. The coefficient values of the proposed fatigue-life prediction model can be used to predict the fatigue life depending on the age of the asphalt and its moisture damage after aging. The degree of fatigue damage can be predicted by calculating the tensile strain using the regression equation and elastic modulus according to the aging and moisture damage.

Compaction Quality Assurance Specifications of Unbound Materials

Compaction quality control/assurance of unbound geomaterials is one of the crucial components in pavement and embankment construction to ensure their performance, stability, and sustainability. Conventional density-based methods such as nuclear density gauge to determine the compaction quality have been widely used due to the straightforward relationship between the readings and targeted material property. Recent modifications in construction standards and the introduction of the Mechanistic-Empirical Pavement Design Guide have inspired a growing interest in developing and implementing strength/stiffness-based compaction control quality assurance (QA) specifications. Numerous studies have been dedicated to investigating the efficiency and effectiveness of the stiffness-based compaction QA tools. This paper presents a comprehensive review of the recent compaction QA relevant literature and surveys. Findings of different approaches for studying QA devices, and the main results of the existing models, experiments, and engineering practices were summarized. Several in situ spot QA technologies, including the latest compaction QA technologies [e.g., the lightweight deflectometer (LWD)], were highlighted, and their efficiency and effectiveness were compared. The review also summarized the intercorrelations between different devices, the correlations between in situ QA device readings and mechanical properties of unbound material, findings of the numerical simulations, and case studies and current practices using different QA tools. The recommendations for future research needs and practical implementations were identified and discussed.

Predicting Flexible Pavement Distress and IRI Considering Subgrade Resilient Modulus of Fine-Grained Soils Using MEPDG.

This paper highlights the subgrade resilient modulus (MR), which is recognized as an important parameter to characterize the stiffness of the subgrade soil for designing flexible pavement. In this study, 18 thin-walled Shelby tube samples of fine-grained subgrade soils were collected from two sites in South Carolina (Laurens/SC-72 and Pickens/SC-93) and tested in the laboratory using AASHTO T307-99 to obtain the MR. In addition, falling weight deflectometer (FWD) tests were performed on the same pavement sections to obtain the back-calculated MR(FWD) per the AASHTOWare 2017 back-calculation tool. A subgrade MR catalog was established and used to select hierarchical Input Level 2 for Pavement Mechanistic-Empirical design (PMED) analysis (v 2.6.1). The PMED analysis was run for 20 years. The Mechanistic-Empirical Pavement Design Guide (MEPDG) and global calibration values were used to predict asphalt concrete (AC) pavement distresses (e.g., rutting, bottom-up fatigue, top-down fatigue, and transverse cracking) and International Roughness Index (IRI) for each pavement section. The predicted values were compared to the field-measured values to determine bias and the standard error of the estimate to validate each distress prediction model for local calibration.

Designing and Building an Intelligent Pavement Management System for Urban Road Networks

Pavement maintenance plays a significant role in megacities. Managing complaints and scheduling road reviews are the two maintenance concerns under the intelligent pavement management system (PMS) plan. In contrast, if the damages are not treated immediately, they will increase over time. By leveraging accurate data from sensors, smart PMS will improve management capability, support sustainability, and drive economic growth in the road network. This research aimed to elaborate on the different modules of an intelligent city pavement network to advance to a sustainable city. First, a 3D mobile light detection and ranging (LiDAR) sensor, accompanied by a camera, was applied as the data collection tool. Although 3D mobile LiDAR data have gained popularity, they lack precise detection of pavement distresses, including cracks. As a result, utilizing RGB imaging may help to detect distresses properly. Two approaches were integrated alongside conducting the data analysis in this paper: (1) ArcGIS pro, developed by Esri Inc., which includes noise removal, digital elevation model (DEM) generation, and pavement and building footprint extraction; (2) the Mechanistic-Empirical Pavement Design Guide (AASHTOWare PMED), which was used to assess site specifications such as traffic, weather, subbase, and current pavement conditions in an effort to design the most appropriate pavement for each road section. For the 3D visualization module, CityEngine (a software from Esri) was used to provide the 3D city model. After implementing the research methodology, we drew the following conclusions: (1) using the AASHTOWare PMED method to make decisions about road maintenance and rehabilitation(M&R) actions can significantly speed up the decision-making process, essentially saving time and money and shortening the project’s duration; and (2) if the road conditions are similar, the smart geographical information system (GIS)-based PMS can make consistent decisions about road M&R strategies, i.e., the interference from human factors is less significant.

Evaluation and Improvement of the Current CRCP Pavement Design Method

Continuously reinforced concrete pavement (CRCP) is a representative type of concrete pavement constructed with continuous steel bars without intermediate transverse expansions. With reference to pavement conditions, CRCP is an exceptional type of concrete pavement according to the Highway Pavement Condition Index (HPCI) and International Roughness Index (IRI). The two main design methods for CRCP are AASHTO 86/93 and the Mechanistic-Empirical Pavement Design Guide (MEPDG). Because of limitations of the AASHTO 86/93 design method, the MEPDG method is more reliable. While incorporating the interactions among geometrics, pavement structure layers, material properties, subgrade, traffic, and environmental conditions, and the prediction values according to the MEPDG method, it matched the measured results of crack spacing and crack width. The MEPDG punchout, crack width and spacing, and load transfer efficiency (LTE) models were evaluated, and results were compared with the test sections in three European countries consisting of different construction details, which were investigated and recorded between 2019 and 2021. In this sense, a calculation tool was developed to consider the influence of different parameters in design process. In addition, sensitivity analyses were executed for the development of punchout, considering various input parameters. The track surveying and the evaluation of the results indicated that the design process can be improved with consideration of some criteria such as crack formation time or adjustment of the correlation between crack width and crack spacing. Due to the very important function of erosion and resulting pumping in the deterioration of CRCP, it is advisable to include the influence of the base layer and the influence of different shoulder type and heavy traffic volume or effect of deflection in the calculations.

Implementation of Mechanistic-Empirical Pavement Design Guide against Indonesian Conditions using Arizona Calibration

Background: As a transportation infrastructure that connects one area to another, roads have an essential role in economic and social growth, together with establishing a location that may improve the quality of life in the surrounding community. For this reason, it is necessary to perform road maintenance when the structural or functional capacity of the road is inadequate, one of which is by overlaying the road. Objective: The main objective of this research is to determine the thickness of the flexible pavement overlay and subsequently examine the damage model produced by the 2015 MEPDG method with Arizona calibration. This study also proposes recommendations for implementing the 2015 MEPDG procedures in Indonesian settings. Methods: The thickness of the road overlay can be designed through a mechanistic-empirical approach, which is commonly referred to as the Mechanistic-Empirical Pavement Design Guide (MEPDG). The back calculation on the BAKFAA program was utilized to examine the existing situation. At the same time, a stress-strain analysis was performed using the KENPAVE software to calculate the response of the pavement structure. Results: The 2015 MEPDG with Arizona calibration by controlling fatigue cracking has resulted in an overlay thickness of 180 mm. In addition, the damage model was obtained for each type of road failure and can be beneficial in estimating the future IRI value. Conclusion: The damage model generated from the 2015 MEPDG procedure is specific to the types of road damage, which can eventually be utilized to predict the future IRI value. It also encompasses local and global calibration variables, such as adjustment factors for its implementation in road pavement conditions in Indonesia. The MEPDG application can be simplified by shortening the mechanistic analysis process, along with reducing traffic variations and the level of detail in the daily climate analysis.

An Efficient and Explainable Ensemble Learning Model for Asphalt Pavement Condition Prediction Based on LTPP Dataset

Accurate prediction of asphalt pavement condition is important to guide pavement maintenance practices. The existing models for pavement condition predictions are predominantly based on linear regressions or simple machine learning techniques. However, additional work on these models is needed to improve their basic assumptions, training efficiency, and interpretability. To this end, a new modeling approach is proposed in this manuscript, which includes a ThunderGBM-based ensemble learning model, coupled with the Shapley Additive Explanation (SHAP) method, to predict the International Roughness Index (IRI) of asphalt pavements. The SHAP method was applied to interpret the underlying influencing factors and their interactions. Twenty features were initially identified as the model inputs, and 2,699 observations were extracted from the Long-Term Pavement Performance (LTPP) database. Three benchmark models, namely the Mechanistic-Empirical Pavement Design Guide (MEPDG) model, the ANN model and the RF model, were used for comparison. The results showed that the developed model achieved a satisfactory result with a R-squared ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathbf {R}^{2}$ </tex-math></inline-formula> ) value of 0.88 and Root Mean Square Error (RMSE) of 0.08, both better than three benchmark models. It ran 86 times and 2.3 times faster than the ANN and RF model, respectively. Feature interpretation was performed to identify the top influencing factors of IRI. The 20-feature model was further simplified based on the analysis result. The simplified model only required six features to efficiently and effectively predict IRI using the proposed ThunderGBM-based approach, which can reduce the workload in data collection and management for pavement engineers.