Search

A novel weighted multi-output neural network (NN) model is proposed for predicting the deterioration of rigid pavements based on Iowa pavement management system data. This first-of-a-kind model simultaneously predicts four pavement condition metrics concerning rigid pavements, including IRI, faulting, longitudinal crack and transverse crack. It provides an opportunity to efficiently evaluate pavement conditions and to make treatment decisions based on multi-condition metrics, such as the pavement condition index (PCI) for budget allocation models. Compared to traditional single-output NN models, this multi-output model is capable of incorporating correlations among different condition metrics. During model training, each condition metric is assigned a weight to reflect its relative importance. When the weights equal to those in the formula for the multi-condition metric, the prediction performance for PCI is optimal (13% lower MSE than optimal, single-output models). The multi-output model improves on the prediction performance for three of the four individual condition metrics compared to optimal single-output models. Results show that the consideration of correlations could improve the prediction performance for single and multi-condition metrics. Finally, variable weighting is critical for achieving the optimal balance of prediction performance among the various metrics as dictated by the needs of the decisionmaker.

The expansion of road networks in emerging economies such as China causes significant greenhouse gas (GHG) emissions. This development is conflicting with China's commitment to achieve carbon neutrality. Thus, there is a need to better understand life cycle emissions of road infrastructure and opportunities to mitigate these emissions. Existing impact studies of roads in developing countries do not address recycled materials, improved pavement maintenance, or pavement-vehicle interaction and electric vehicle (EV) adoption. Combining firsthand information from Chinese road construction engineers with publicly available data, this paper estimates a comprehensive account of GHG emissions of the road pavement network to be constructed in the next ten years in the Shandong province in Northern China. Further, we estimate the potential of GHG emission reductions achievable under three scenario sets: maintenance optimization, alternative pavement material replacement, and EV adoption. Results show that the life cycle GHG emissions of highways and Class 1-4 roads to be constructed in the next 10 years amount to 147 Mt CO2-eq. Considering the use phase in our model reveals that it is the dominant stage in terms of emissions, largely due to pavement-vehicle interaction. Vehicle electrification can only moderately mitigate these emissions. Other stages, such as materials production and road maintenance and rehabilitation, contribute substantially to GHG emissions as well, highlighting the importance of optimizing the management of these stages. Surprisingly, longer, not shorter maintenance intervals, yield significant emission reductions. Another counter-intuitive finding is that thicker and more material-intensive pavement surfaces cause lower emissions overall. Taken together, optimal maintenance and rehabilitation schedules, alternative material use, and vehicle electrification provide GHG reduction potentials of 11%, 4%-16% and 2%-6%, respectively.

This paper proposes a new probabilistic treatment path dependence (PTPD) model for budget allocation within pavement network management problems. During the evaluation of treatment alternatives for a segment in a pavement network, the model considers benefits of both the evaluated treatment and its following actions. It also incorporates the influence of treatment cost and deterioration uncertainties. Treatments are selected for each segment in the pavement network using a risk-based optimization model. Three case studies are presented to illustrate the application and benefits of this new model. Results of the first two cases show that the risk-aversion coefficient in the model influences segment-level treatment selections and pavement network performance. The third case shows that the PTPD model performs better than a benefit-cost ratio model due to the incorporation of uncertainties and treatment path dependence. To obtain a similar performance level, the conventional model requires a 10.4% higher annual budget for the given case study. The results presented here suggest that elements of this model – notably consideration of uncertainty in deterioration and cost, treatment path dependency, and explicit risk trade-offs – could be incorporated into asset management tools to improve the cost-effectiveness of pavement network planning.

Life-cycle cost analysis (LCCA) is a way to evaluate the long-term cost effectiveness of different pavement designs or treatment actions. Owing to the existence of uncertainties, many probabilistic LCCA models have been proposed. They mainly use a prescribed treatment schedule or determine schedules by mechanistic-empirical analysis, potentially leading to the overestimation of life-cycle cost (LCC). In this paper, a new probabilistic simulation-optimization LCCA model is proposed. This new model determines treatment schedules by minimizing total LCC, including agency and user cost, which is different from current probabilistic models. In addition, it also incorporates uncertainties of treatment costs and deterioration processes. Two case studies are presented. The first one shows the influence of treatment schedule uncertainties on LCC distributions. After considering treatment schedule uncertainties, a tighter LCC distribution is estimated. The second case study compares the new model and a conventional prescribed-schedule model from the perspective of pavement design selection. The results show that the simulation-optimization model could lead to different preferred pavement designs than the prescribed-schedule model.

Concrete is a critical component of deep decarbonization efforts because of both the scale of the industry and because of how its use impacts the building, transportation, and industrial sectors. We use a bottom-up model of current and future building and pavement stocks and construction in the United States to contextualize the role of concrete in greenhouse gas (GHG) reductions strategies under projected and ambitious scenarios, including embodied and use phases of the structures’ life cycle. We show that projected improvements in the building sector result in a reduction of 49% of GHG emissions in 2050 relative to 2016 levels, whereas ambitious improvements result in a 57% reduction in 2050, which is 22.5 Gt cumulative saving. The pavements sector shows a larger difference between the two scenarios with a 14% reduction of GHG emissions for projected improvements and a 65% reduction under the ambitious scenario, which is ∼1.35 Gt. This reduction occurs despite the fact that concrete usage in 2050 in the ambitious scenario is over three times that of the projected scenario because of the ways in which concrete lowers use phase emissions. Over 70% of future emissions from new construction are from the use phase.

Performance-based planning is an important tool for allocating treatment resources across a pavement network from a set of candidate treatments with a budget constraint. Existing research focuses on improving allocation decisions through changes in the optimization algorithm without considering the consequences of how optimization analyses are framed. In this paper, both environmental and economic performance is evaluated for different problem framing in the form of different treatment strategies that consist of treatment materials, treatment types, and evaluation period. Results show that the proposed strategy that uses both concrete and asphalt, different treatment types, and a long evaluation period could reduce GHG emissions and improve pavement network performance based on the Iowa U.S. route network. Compared to a conventional 5-year asphalt-only strategy, proposed strategy can accomplish this with an annual budget that is 32% smaller and reduce associated GHG emissions by 21%. These results contribute to achieving a sustainable pavement network.

Electrification and lightweighting technologies are important components of greenhouse gas (GHG) emission reduction strategies for light-duty vehicles. Assessments of GHG emissions from light-duty vehicles should take a cradle-to-grave life cycle perspective and capture important regional effects. We report the first regionally explicit (county-level) life cycle assessment of the use of lightweighting and electrification for light-duty vehicles in the U.S. Regional differences in climate, electric grid burdens, and driving patterns compound to produce significant regional heterogeneity in the GHG benefits of electrification. We show that lightweighting further accentuates these regional differences. In fact, for the midsized cars considered in our analysis, model results suggest that aluminum lightweight vehicles with a combustion engine would have similar emissions to hybrid electric vehicles (HEVs) in about 25% of the counties in the US and lower than battery electric vehicles (BEVs) in 20% of counties. The results highlight the need for a portfolio of fuel efficient offerings to recognize the heterogeneity of regional climate, electric grid burdens, and driving patterns.

The impacts of climate conditions, the geometry of concrete in use, and end-of-life and surface treatment actions have received little attention in previous studies of carbon uptake calculation. This work attempts to advance the knowledge of pavement carbon uptake based on a bottom-up approach and using state-level data to estimate the carbon uptake of the US pavements and its associated cost. To do so, a pavement management system model was developed to predict the performance and characteristics of the national pavement network using local maintenance and repair practices and data science approaches. Then, a high-resolution carbon uptake estimation method was applied to this system model. Our results show that 5.8 million metric tons (Mt) CO2 can be sequestered by the US pavement network, of which 52% will be sequestered when the demolished concrete is stockpiled at the end of life. The climate condition and mix design practices in California result in this state having the largest carbon uptake during the use phase of the pavement life cycle, while the end-of-life uptake is larger in those states that have an extensive composite pavement network. Extending the stockpiling period up to 30 years can increase the total end-of-life carbon uptake to 11.8 Mt CO2. The cost associated with concrete stockpiling varies depending on the scenario, but abatement costs for a 1 to 17-year stockpiling timeframe are $25–100/ton CO2, making end-of-life stockpiling a competitive strategy for greenhouse gas mitigation of the road network.

When concrete is exposed to fire, spalling may occur. As regards this phenomenon, many numerical models have been proposed and developed during the last 30 years. It is state-of-the-art that the governing equations for consideration of spalling of concrete, which represent the basis of such models, are fully coupled, accounting for several important physical and chemical phenomena of different significance. The number of material parameters has increased with increasing complexity of these equations. This suggests exploring the relative significance of these parameters. For that purpose, a one-dimensional mathematical model for the analysis of concrete exposed to fire is used. By means of exploring the sensitivity of the temperature and the gas pressure at a specific point in a concrete floor slab to variations of several material parameters in the employed constitutive relations, the goal of this work is achieved.

The critical heat flux (CHF) corresponding to the departure from nucleate boiling (DNB) crisis is essential to the design and safety of a two-phase flow boiling system. Despite the abundance of predictive tools available to the thermal engineering community, the path for an accurate, robust CHF model remains elusive due to lack of consensus on the DNB triggering mechanism. This work aims to apply a physics-informed machine learning (ML)-aided hybrid framework to achieve superior predictive capabilities. Such a hybrid approach takes advantage of existing understanding in the field of interest (i.e., domain knowledge) and uses ML to capture undiscovered information from the mismatch between the actual and domain knowledge-predicted target. A detailed case study is carried out with an extensive DNB-specific CHF database to demonstrate (1) the improved performance of the hybrid approach as compared to traditional domain knowledge-based models, and (2) the hybrid model's superior generalization capabilities over standalone ML methods across a wide range of flow conditions. The hybrid framework could also readily extend its applicability domain and complexity on the fly, showing an elevated level of flexibility and robustness. Based on the case study conclusions, the window-type extrapolation mapping methodology is further proposed to better inform high-cost experimental work., United States. Department of Energy (Contract DE-AC05-00OR22725), Consortium for Advanced Simulation of Light Water Reactors