Dimitrios Goulias

With the rapid deterioration of infrastructure, accurate predictions of bridge deck performance is critical in bridge management systems (BMS). The US Federal Highway Administration (FHWA) has developed the National Bridge Inventory (NBI) database, which has accumulated a vast repository of bridge performance data in the US. Although alternative methods have been used for harvesting such data, the use of machine learning (ML) has been explored to a lesser degree due to its modeling complexity. Objective of this study was to develop and assess alternative sequential models by employing long short-term memory (LSTM) and convolutional neural networks (CNN). The advantage of such modeling alternatives in relation to past ML studies is that consider (1) fixed or variable number of sequences of observations combined with (2) continuous or flexible sequences, producing better condition rating predictions. The results indicated that the models exceeded accuracy prediction from the mid 80% in past studies to levels of 95% for Maryland bridges, and all the way to 97% when Massachusetts data were added. Such results indicate that the proposed models are able to better capture the complex relationships between bridge-related features and condition rating. Transferability of the proposed approach was confirmed with the successful model response when data from the second state were used. It is expected that improved predictions of condition rating will promote more effective maintenance and rehabilitation strategies in BMS.

US Transportation Collection2020PDFTech ReportGoulias, DimitriosUniversity of Maryland (College Park, Md.). Department of Civil and Environmental EngineeringMaryland. Dept. of TransportationMaryland. State Highway Administration. Office of Policy & ResearchUnited States. Federal Highway Administration. State Planning and Research Program (SPR)Maryland. State Highway Administration. Office of Policy & ResearchMarylandBituminous overlaysBridge decksCondition surveysData collectionDelaminationGround penetrating radarMD-20-SHA/UM/4-54SHA/UM/4-54SPR-B Final Report (December 2017-December 2019)Over the past several years Maryland Department of Transportation State Highway Administration (MDOT SHA) developed a Ground Penetration Radar (GPR) data collection plan for bridge decks. GPR data was collected and analyzed to monitor several hundred bridge decks. MDOT SHA worked with the Maryland Environmental Services (MES) and the University of Maryland (UMD) to develop new analysis modules for concrete delamination and HMA overlay condition and evaluate the feasibility of higher-speed protocols for SF-GPR data collection. A bridge deck condition assessment model (BDCAM) was developed to estimate the deck condition and condition state. Deck condition is defined based on a fuzzy model of the various levels of defect and deterioration of the deck. The UMD study concluded that the BDCAM model estimates agree with the NBI values for 90.9% of the 219 bridge decks analyzed within two levels of the condition scale. The comparison of BDCAM analysis with state inspection deck reports for eight bridges provided consistent conclusions for seven out of the eight cases, all in the \u201cfair\u201d category. The study also concluded that it is possible to increase the GPR data acquisition speed from 10 mph to 13 mph on driving lanes with low surface roughness (IRI less than 100).110

PDFBriefSHA/UM/3-11Ground penetrating radarStructural health monitoringCondition surveysPavement management systemsPrecast concreteBridge decksQuality controlMarylandMaryland. State Highway AdministrationGoulias, DimitriosScott, MichaelMaryland. State Highway AdministrationUS Transportation CollectionRecently Maryland State Highway Administration (SHA) started to explore use of GroundPenetrating Radar (GPR) technology to provide quantitative information for improveddecision making and reduced operating costs. To take full advantage of the GPRcapabilities, improved analysis techniques need to be developed and implemented

The project described in this paper undertook an extensive investigation on the use of reinforced recycled-based resins for developing alternative poles for highway utility poles. The study built on an initial feasibility analysis for examining the material behavior and performance of the reinforced resins and fine tuning alternative pole designs. The project objective was to identify a cost effective alternative to wood poles using plastic resins. The project stages included a preliminary technical and environmental feasibility analysis, mechanical testing of resins, creep evaluation, pole design analysis, quality control testing, and field demonstration with a prototype pole for long term monitoring.

Under TE-30, High Performance Concrete Pavement program, several states are undertaking a variety of innovative research in high performance concrete pavement materials and innovative design/construction features. This project addressed the needs of the Maryland State Highway Administration (MSHA) in exploring the use of fiber reinforced and low shrinkage concrete in pavements. The objective of this study was to examine the design and laboratory performance of these materials for Maryland conditions, monitor their laboratory and field performance, and quantify potential benefits. MSHA constructed pavement test sections with control, fiber reinforced, and low shrinkage concrete on the Route 50 Salisbury Bypass. The test sections are designed to test the pavement's durability in terms of its resistance to cracking. Two approaches were studied: low-shrinkage concrete and fatigue-resistant concrete. This is also a significant project in that high-performance concrete is being used on a large pavement project, not just for bridge decks and other specialized structures. The study included both laboratory evaluation of these mixtures and field performance through on site instrumentation and analytical evaluation. The deflection and strain data were used in the analytical evaluation. This analysis indicated that the best estimates of k and Ec both from the measured deflections and measured strains are k of 350 pci and Ec of 5,000,000 psi. These are both reasonable values for the embankment soils and concrete conditions at the site at the time of the load tests.

State highway agencies (SHAs) use contractor test results in the acceptance and pay rewards of materials and construction. Fand ttests are commonly used for validating contractor test results. However, these tests are based on sample statistics, which may lead to significant acceptance risks and inconsistent rewards. In this study a Monte Carlo simulation process was developed to systematically quantify the acceptance risks and assess the implications on pay factors (PF). The simulation was performed using typical acceptance quality characteristics (AQCs), such as thickness, for Portland cement concrete (PCC) pavements. The statistical power of the Fand t-test was determined. The analysis indicated that specific combinations of contractor and agency sample sizes and population characteristics have a greater impact on the acceptance risks and may provide inconsistent PF. Thus, the findings of this study can assist both agencies and producers to better assess the acceptance risks and rewards associated with the validation procedures. Therefore, the proposed methodology can be adopted by SHAs to develop statistically valid verification procedures and thus more rational quality assurance (QA). Producers may use such analysis to identify the level of risks and rewards associated with the current production and identify potential improvements in quality.

for developing high performance concrete pavements. The paper also provides an overview of currently ongoing activities in the U.S. materials for pavement structures is presented along with the expected benefits. mixtures. Specifically, the design, behavior and performance of concrete benefits on pavement performance fiom the use of high performance concrete pavements. This paper presents current efforts in this area and the expected accelerated and systematic studies for developing high performance concrete concrete pavement construction in the U.S. generated the need for carrying out methods, and ii) construction techniques. Recently, significant emphasis on around the world are searching for improved i) pavement materials and design use of longer lasting materials and pavement structures.Thus, highway agencies The increasing loading conditions on highway pavements requires the 1. High performance concrete pavements in the U.S.

Production of concrete is consuming significant amounts of raw materials, high level of energy, and is heavily dependent on fossil fuels. This induces significant harmful impact on the environment. Scientific research is being conducted worldwide on the possibility of using different waste by-products in the production of concrete, particularly as a substitute for cement clinker. Rock dust of different geological origin can be considered as potential alternative material that can be used in cement composites production. However, there are some conflicting findings concerning the effect of rock dusts as partial cement replacement on the physical and mechanical properties and durability of cement composites as reported in the literature. Thus, a comprehensive assessment and analysis are needed to evaluate the value of rock dust application as cement replacement in concrete production. This paper presents a comprehensive review of the findings from scientific articles concerning the use of rock dust of different geological origins in mortar and concrete productions. The effect of rock dusts as partial cement substitution on cementitious composites properties were analyzed particularly on the cement hydration, the concrete and mortar mixture properties, mechanical properties and durability. The impact of rock dust is mainly related to the filler effect i.e., due to modification of particle size distribution, heterogenious nucleation, and cement dilution. This effect is more significant when cement is substituted with a rock dust of greater fineness than cement. Partial replacement of cement with up to about 10-15% of rock powder does not deteriorate cement composite properties. However higher substitution leads to reduction of mechanical properties and cement composite durability decline.

Production of concrete is consuming significant amounts of raw materials, high level of energy, and is heavily dependent on fossil fuels. This induces significant harmful impact on the environment. Scientific research is being conducted worldwide on the possibility of using different waste by-products in the production of concrete, particularly as a substitute for cement clinker. Rock dust of different geological origin can be considered as potential alternative material that can be used in cement composites production. However, there are some conflicting findings concerning the effect of rock dusts as partial cement replacement on the physical and mechanical properties and durability of cement composites as reported in the literature. Thus, a comprehensive assessment and analysis are needed to evaluate the value of rock dust application as cement replacement in concrete production. This paper presents a comprehensive review of the findings from scientific articles concerning the use of rock dust of different geological origins in mortar and concrete productions. The effect of rock dusts as partial cement substitution on cementitious composites properties were analyzed particularly on the cement hydration, the concrete and mortar mixture properties, mechanical properties and durability. The impact of rock dust is mainly related to the filler effect i.e., due to modification of particle size distribution, heterogenious nucleation, and cement dilution. This effect is more significant when cement is substituted with a rock dust of greater fineness than cement. Partial replacement of cement with up to about 10-15% of rock powder does not deteriorate cement composite properties. However higher substitution leads to reduction of mechanical properties and cement composite durability decline.

With the rapid deterioration of infrastructure, accurate predictions of bridge deck performance is critical in bridge management systems (BMS). The US Federal Highway Administration (FHWA) has developed the National Bridge Inventory (NBI) database, which has accumulated a vast repository of bridge performance data in the US. Although alternative methods have been used for harvesting such data, the use of machine learning (ML) has been explored to a lesser degree due to its modeling complexity. Objective of this study was to develop and assess alternative sequential models by employing long short-term memory (LSTM) and convolutional neural networks (CNN). The advantage of such modeling alternatives in relation to past ML studies is that consider (1) fixed or variable number of sequences of observations combined with (2) continuous or flexible sequences, producing better condition rating predictions. The results indicated that the models exceeded accuracy prediction from the mid 80% in past studies to levels of 95% for Maryland bridges, and all the way to 97% when Massachusetts data were added. Such results indicate that the proposed models are able to better capture the complex relationships between bridge-related features and condition rating. Transferability of the proposed approach was confirmed with the successful model response when data from the second state were used. It is expected that improved predictions of condition rating will promote more effective maintenance and rehabilitation strategies in BMS.

State highway agencies (SHAs) use contractor test results in the acceptance and pay rewards of materials and construction. Fand ttests are commonly used for validating contractor test results. However, these tests are based on sample statistics, which may lead to significant acceptance risks and inconsistent rewards. In this study a Monte Carlo simulation process was developed to systematically quantify the acceptance risks and assess the implications on pay factors (PF). The simulation was performed using typical acceptance quality characteristics (AQCs), such as thickness, for Portland cement concrete (PCC) pavements. The statistical power of the Fand t-test was determined. The analysis indicated that specific combinations of contractor and agency sample sizes and population characteristics have a greater impact on the acceptance risks and may provide inconsistent PF. Thus, the findings of this study can assist both agencies and producers to better assess the acceptance risks and rewards associated with the validation procedures. Therefore, the proposed methodology can be adopted by SHAs to develop statistically valid verification procedures and thus more rational quality assurance (QA). Producers may use such analysis to identify the level of risks and rewards associated with the current production and identify potential improvements in quality.

Assessing concrete quality as construction goes on provides early warnings of potential flaws and leads to timely corrections in mix proportioning and placement techniques. Compressive strength and maturity modeling are among the most common parameters used by the concrete industry. Past studies indicated that non-destructive methods, NDTs, relate well to maturity and concrete strength predictions. In this study, the hydration temperature–time history of concrete was explored in defining “master curves” for concrete maturity for the first time. Well-accepted NDTs, such as ultrasonic pulse velocity and resonant frequency, were used in this effort. The study findings indicated that the novel approach of “master curves” for the maturity of concrete can be defined and follow a generalized logarithmic form. The best fit models relating NDT response and the maturity temperature–time product provided a high coefficient of determination (i.e., in almost all cases above 0.9 and p < 0.05),...

Highway agencies monitor the condition of thousands of bridge decks every year. Even though Ground Penetrating Radar (GPR) has been used in bridge-deck evaluation, Step-Frequency GPR (SF-GPR) provides advanced condition assessment yet requires extensive and complex post-processing analysis. An SF-GPR analysis system was recently developed and used for monitoring the condition of all the bridge decks in the state of Maryland. The objective of this study was to develop a bridge deck condition rating approach using fuzzy sets modeling on the SF-GPR data and analysis. The fuzzy sets membership functions needed to reflect rating score categories similar to those considered in the National Bridge Inventory (NBI) database for uniformity. Thus, the fuzzy sets modeling was built considering nine condition membership functions. The overall bridge deck condition score leading to each of the nine condition states was based on both physical and condition-related bridge deck parameters as obtaine...

Due to the depletion of natural aggregates and high maintenance cost of highway systems, developing sustainable asphalt concrete (AC) mixes that use waste materials instead of virgin raw materials is necessary. A large amount of waste glass material is globally generated per year that could be beneficial to sustain the asphalt industry. In this context, the present paper evaluates the properties and performance of AC mixtures that utilize crushed waste glass as a replacement material of filler aggregates. Three AC mixes with percentages of filler replacement in the range from 0%, 25%, to 50% were fabricated. Complex modulus testing was performed to evaluate the dynamic modulus |E*| and phase angle δ over a range of temperatures and loading frequencies. In addition, the flow number (FN) test was conducted to assess the rutting potential of the mixtures. The results showed that the mix containing 25% of crushed glass is likely to better resist fatigue cracking; however, the inclusion ...

The use of Geographic Information Systems in transportation (GIS-T) can greatly enhance spatial and time dependent analysis. Its use in highway engineering and transportation planning/operations has been significantly intensified since GIS permits the assimilation, integration, modeling, and visualization of time and space related data and predictions. This paper presents two representative case studies on the development and use of GIS-T in: i) pavement management and highway analysis, integrating performance models, user and agency cost models and pavement condition information (in terms of inventory and maintenance data); and ii) emergency management and traffic operations, integrating models of human behavior, transportation network characteristics, and detailed regional demographic and land-use information. In the development of these GIS based management systems, temporal and spatial modeling was used for describing spatio-temporal phenomena and predicting future conditions. T...

With the increasing cost of infrastructure monitoring and maintenance and the limited funding for these activities the engineering community is looking towards the development of smart materials. Today there are several examples of innovative smart materials requiring however a) increased hightech knowledge for their installation monitoring and interpretation of data, and b) significant expense for their installation in large scale projects, such as highways and bridges. This paper presents the result of a project undertaken for investigating the development of simple and low cost sensor for concrete and asphalt mixtures. Several configurations of metallic grids were imbedded into concrete and asphalt beams that were exposed to different stress and strain levels, and temperature differentials. The input and output voltage of these grid sensors were monitored in relation to the applied external stimuli. The results indicated that such a simple and low cost sensing devices can be used...

Construction and road maintenance are ordinary activities frequently undertaken to maintain the well-being of road infrastructure. When these activities take place, they impact traffic flow and generate delays to the road users. Costs are imposed on the users on heavily traveled routes due to traffic slowdowns or even shutdowns. At rush-hour these direct and indirect costs come to a peak. Construction and road maintenance closures can take place at times when the negative impacts would be minimized. This study focused on the appropriate guidelines for lane occupancy charges that would eventually minimize the disutility of traffic lane closure. The project research team examined heavily traveled locations in the New Jersey region, with the cooperation of New Jersey Department of Transportation engineers. They examined traffic and construction patterns to be used in the analysis and developed general occupancy charge guidelines. Information regarding traffic flow with respect to time ...

The New York City Department of Transportation (NYCDOT) Parking Division is responsible for the largest parking meter network in the U.S. This study addressed the need of the NYCDOT Parking Division to review its current operations and evaluate operability measures. The study was undertaken in coordination with the research unit, the maintenance and revenue collection units, and the computer services unit. The objective of the study was to examine and evaluate the current definition of operability, review meter inspection/maintenance and revenue collection operations, identify improvements, and provide a systematic method of implementing these improvements to increase efficiency, and ultimately, revenue. The study was organized in two phases. In the first phase, the current practice was reviewed using one-year historical data for identifying potential improvements in meter inspection/maintenance and revenue collection operations. In the second phase, testing of the assumptions and f...

Significant investigation is being conducted on the use of crumb rubber or tire chips in asphalt mixtures either as binder enhancement or aggregate replacement. However, its use in Portland cement concrete was not investigated to the same degree. For Portland cement concrete (PCC), rubber from granulated tires may be used as an elastic aggregate modifying the brittle failure of concrete and increasing its ability to absorb higher amounts of energy before failure. This paper presents the results of an experimental study conducted for assessing the feasibility of using recycled tire rubber in PCC. Various rubber contents were considered in the design of several Portland cement concrete mixtures, including control mixtures with no rubber. The concrete samples were evaluated with the fundamental resonant frequency method, a well accepted non-destructive testing technique and the results were then coupled with destructive methods for relevant comparisons. The cost effectiveness of these ...

Pavement data collection is the most expensive and time consuming component of Pavement Management System (PMS). Thus, possible methods of minimizing the need of such data might be critical in reducing pavement condition monitoring costs. Also the ability to relate pavement performance prediction models (frequently roughness based) to hot mix asphalt field performance models (distress based) provides valuable conclusions and input in pavement design, performance assessment, maintenance and rehabilitation strategies. Objective of this study was to examine whether specific distresses can influence roadway profile so as to be able to relate the two. The influence of pavement distresses on road profile has been investigated over the years. However, past studies provided conflicting conclusions. Thus, in this study an alternative approach was considered due to the availability of high quality and detailed distress data collected with the Laser Crack Measurement System (LCMS) of the Automatic Road Analyzer, ARAN. As it was expected, specific distresses have higher impact in longitudinal roughness since they are present on the roadway surface at regular intervals (i.e., specific frequencies). For this reason, instead of using summary indexes (i.e., International Roughness Index (IRI), Pavement Condition Index (PCI)), the Power Spectral Density (PSD) of the roadway profile at specific frequency bandwidths was considered along with distresses. The analysis indicated that a specific subset of distresses is affecting roughness at definite wavelength frequencies. Alligator cracking and rutting standard deviation provided the best correlation. IRI was correlated better with distress (e.g. rutting standard deviation) at lower profile frequencies. At high frequency domain (i.e., below 0.8 m wavelengths) better correlation between IRI and high severity cracking was observed through the PSD. Considering multiple frequencies in the regression models between roughness and distresses, the goodness of fit has not necessarily improved. However, the role of different bandwidths was evident. In addition to the specific results, the methodology presented in this study can be used elsewhere to assess potential relations between pavement roughness and distress components.

The use of recycled materials and industrial by-products in pavement construction and rehabilitation can achieve substantial benefits in saving nature resources and reducing energy consumption as well as greenhouse gas (GHG) emissions. Alternative geological origin rock dust for the partial replacement of fine aggregate and/or cement in Portland cement concrete (PCC) pavements may provide positive environmental and economic benefits. The objective of this study was to quantitatively assess the life-cycle economic and environmental impacts when rock dust is used in PCC pavement roadway construction. Previous studies have primarily focused on the economics and/or environmental impacts during the material production process. Thus, a methodological framework considering all stages (such as material production, transportation, construction, maintenance, rehabilitation and end of life), involved in the life-cycle assessment of concrete pavements is proposed when using recycled materials/b...

Background: Highway agencies explore the use of non-destructive testing (NDT) for assessing the condition of their infrastructure in the most cost-effective manner. NDTs can provide as-built construction quality, as well as in service condition assessment especially for older structures where some critical information and/or maintenance records are often not available. In many cases, the use of NDTs requires "ground true" data for either calibration or validation. Surveys for obtaining such information is often challenging for in service structures, time consuming and costly.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Pavement data collection is the most expensive and time consuming component of Pavement Management System (PMS). Thus, possible methods of minimizing the need of such data might be critical in reducing pavement condition monitoring costs. Also the ability to relat e pavement performance prediction models (frequently roughness based) to hot mix asphalt field performance models (distress based) provides valuable conclusions and input in pavement design, performance assessment, maintenance and rehabilitation strategies. Objective of this study was to examine wh ether specific distresses can influence roadway profile so as to be able to relate the two. The influence of pavement distresses on road profile has been i nvestigated over the years. However, past studies provided conflicting conclusions. Thus, in this study an alternative approach was considered due to the availability of high quality and detailed distress data collected with the Laser Crack Measurement System (LCMS) of the Automatic Road Analyzer, ARAN. As it was expected, specific distresses have higher impact in longitudinal roughness since they are present on the roadway surface at regular intervals (i.e., specific frequencies). For this reason, instead of using summary indexes (i.e., International Roughness Index (IRI), Pavement Condition Index (PCI)), the Power Spectral Density (PSD) of the roadway profile at specific frequency bandwidths was considered along with distresses. The analysis indicated that a s pecific subset of distresses is affecting roughness at definite wavelength frequencies. Alligator cracking and rutting standard deviation provided the best correlation. IRI was correlated better with distress (e.g. rutting standard deviation) at lower profile frequencies. At high frequency domain (i.e., below 0. 8 m wavelengths) better correlation between IRI and high severity cracking was observed through the PSD. Considering multiple frequencies in the regression models between roughness and distresses, the goodness of fit has not necessarily improved. However, the role of different bandwidths was evident. In addit ion to the specific results, the methodology presented in this study can be used elsewhere to assess potential relations between pavement roughness and distre ss components.

Many state highway agencies (SHAs) have adopted pay adjustment provisions in their acceptance plans for construction and materials. In these payment adjustment acceptance plans, the percentage of material within specification limits (PWL) has been selected as the quality measure to relate production quality to pay factors, and pay equations are used to determine a pay factor for a lot based on PWL. Various pay equations have been proposed by the highway community for adoption in SHAs' specifications. However, the effectiveness of these pay equations has not been fully evaluated. Another issue concerning the pay adjustment acceptance plans is the risk associated with single and multiple pay factors. The purpose of this study was to evaluate the effects of different pay equations commonly used by SHAs and the risks associated with pay adjustment acceptance plans. This was achieved by developing operating characteristic curves associated with various pay factors and expected pay curves and Monte Carlo simulation for assessing the effects in the long run. The methodology suggested in this paper is transferable elsewhere where similar materials and specifications are used for the acceptance of pavements.

Pavement skid resistance has a significant role in traffic accidents, especially in wet conditions. Pavement surface characteristics are affected by both materials and mixture properties. This study explored a 'novel' approach to pavement friction analysis in modelling and relating pavement friction to materials and mixture properties. Structural equation modelling (SEM) takes advantage of the correlation/collinearity among one or more predictor variables in generating predictive models for a response variable. While SEM has been used in a variety of fields, in pavement friction the use of such statistic approach has not been explored, and thus it is a 'novel approach' to pavement friction modelling in relation to the past modelling efforts. Thus, in this study the selection of SEM modelling is advantageous so as to, (i) capture the interdependency of mixture and material variables in hot mix asphalts; and (ii) address the high number of predictor variables in relation to the number of observations (small sample size of observations). While data from Maryland were used in this analysis the methodology can be used elsewhere reflecting similar materials and pavement conditions.

In the current trend of sustainability, the concrete community has been aggressively looking into adopting green construction material practices and at the same time improving concrete quality and performance for extensive service life and adaptive reuse. Concrete is the construction material most used in the world, with an estimated yearly production of 2.35 billion tons worldwide. In the United States, it is estimated that on average approximately 5% of the ready-mix concrete produced is unused and returned to the plant with only a small portion reused. Such a material when further processed, identified as crushed returned concrete aggregate (CCA), has a significant residual value because among other things it is free of contaminants and has better quality than recycled concrete aggregate (RCA). The objective of this study was to assess the shrinkage behavior of CCA concrete mixtures produced with aggregate from returned concrete. The aggregate was prepared from concrete of different strength. The virgin aggregate (stone) was replaced either partially or at 100% level in the concrete mixtures. The response of the hyperbolic shrinkage prediction model was examined, and based on the experimental results an alternative model is proposed. The proposed model and methodology can be used to estimate the drying shrinkage of CCA mixtures, and eventually can be adopted for assessing the shrinkage behavior of these concrete mixtures in other regions.