Turner-Fairbank Highway Research Center Research Articles

Fusion and Visualization of Bridge Deck Nondestructive Evaluation Data via Machine Learning

To maintain infrastructure safety and integrity, nondestructive evaluation (NDE) technologies are often used for detection of subsurface defects and for holistic condition assessment of structures. While the rapid advances in data collection and the diversity of available sensing technologies provide new opportunities, the ability to efficiently process data and combine heterogeneous data sources to make robust decisions remains a challenge. Heterogeneous NDE measurements often conflict with one another and methods to visualize integrated results are often developed ad hoc. In this work, we present a framework to support fusion of multiple NDE techniques in order to improve both detection and quantification accuracy while also improving the visualization of NDE results. For data sources with waveform representations, the discrete wavelet transform (DWT) is used to extract salient features and facilitate fusion with scalar-valued NDE measurements. The description of a signal in terms of its salient features using a wavelet transform allows for capturing the significance of the original data, while suppressing measurement noise. The complete set of measurements is then fused using nonparametric machine learning so as to relax the need for Bayesian assumptions regarding statistical distributions. A novel visualization schema based on classifier confidence intervals is then employed to support holistic visualization and decision making. To validate the capabilities of the proposed methodology, an experimental prototype system was created and tested from NDE measurements of laboratory-scale bridge decks at Turner-Fairbank highway research center (TFHRC). The laboratory decks exhibit various types of artificial defects and several non-destructive tests have already been carried out by research center technicians to characterize the mechanical properties of the materials and the existing damages. The behaviors of several different data fusion approaches are compared here. Based on the presented analyses, it is demonstrated that fusion via support vector machines provided the most robust and consistent data fusion and defect detection capabilities.

Calibrating Communication Simulator for Connected Vehicle Applications

Research efforts focused on the connected vehicle (CV) technology applications typically assume perfect communications among the vehicles and between the vehicles and the roadside equipment. However, a few studies, including this one, pointed out that the wireless communications experience packet drops, which might lead to a serious downgrade of the safety critical CV applications. Thus, the wireless communication simulators used to emulate the communications performance need to be properly calibrated to replicate the real-world vehicular communication environments. This study calibrated an NCTUns simulator for the dedicated short range communications (DSRC) of CV technology using the DSRC field test data executed on an instrumented intersection at the Turner–Fairbank Highway Research Center. Physical layer parameters (e.g., data rate and transmission power) as well as channel models are calibrated. The calibration applied a Latin hypercube sampling technique to generate multiple combinations of parameter sets. The calibrated NCTUns simulator produced much more realistic outputs than the uncalibrated one. Then a signalized intersection was simulated in a case study to further investigate the packet drops of DSRC-based CV communications. The results indicated that there were significant packet drops, requiring further research before implementing safety critical CV applications.

Comparison between fatigue performance of horizontal cores from different asphalt pavement depths and laboratory specimens

This study presents a comparison between the fatigue performance of field cores obtained from different layers (top and bottom) and laboratory produced–laboratory compacted specimens. The field cores were obtained from the Accelerated Pavement Testing (APT) facility at Turner-Fairbank Highway Research Center (TFHRC), located in McLean, Virginia, USA. Uniaxial push-pull (compression-tension) fatigue tests were conducted on samples obtained from the top 50 mm and the bottom 50 mm layers of 100 mm thick asphalt pavements. The samples were obtained by horizontally coring slabs extracted from six APT lanes. The damage characteristic curves (i.e., C versus S curve of the viscoelastic continuum damage (VECD) theory) and the fatigue lives of the field cores were compared with those of the laboratory specimens. The experimental test results and the subsequent analysis findings revealed comparable damage characteristic curves and fatigue lives for the top and bottom field specimens; meanwhile, the field and laboratory specimens exhibited different damage characteristic curves and fatigue lives.

Evaluation of Comprehension and Legibility of International and Domestic Nonstandard Symbol Signs

This paper documents a research project that evaluated proposed new traffic signs to ensure that the signs were effective when human factors requirements were taken into consideration. The following sign messages were selected for symbol development and evaluation: combination horizontal alignment–advisory speed, congestion ahead, cross-street preferential lane warning, do not enter, do not pass, electric-vehicle charging station, fallen rocks, flagger ahead, low-shoulder warning, maximum width, no left turn ahead, railroad crossing on leg of roundabout, road narrows, survey crew, trolley crossing, uneven lanes, and winery. The team conducted 96 surveys to determine driver comprehension and legibility for each alternative sign design. The surveys were administered to the general driving public at the Turner–Fairbank Highway Research Center in McLean, Virginia. On the basis of driver input and the results of the comprehension and legibility testing, the research team provided recommendations on symbols that should be included in the next edition of the Manual on Uniform Traffic Control Devices.

Evaluation of Asphalt Damage and Cracking Development with Seismic Pavement Analyzer

Fatigue cracking in asphalt pavements is an indicator of structural failure, and early detection of fatigue cracking would allow for improved preventive maintenance. The evaluation of cracking relies on the distress's being visible and possibly in advanced stages. Detection of load-related cracking at an early stage or before initiation is greatly needed. The portable seismic property analyzer (PSPA), a nondestructive device that monitors the change in modulus in pavement surface layers, was used in this study to measure the change in pavement stiffness response at the same time as the development of load-associated cracking. PSPA tests were conducted at the FHWA Turner–Fairbank Highway Research Center accelerated loading facility on full-scale test pavements being subjected to repeated wheel loading to induce cracking. PSPA measurements taken at a predetermined number of loading passes were evaluated to find crack initiation or propagation. The patterns for the amount of modulus reduction before and after the initiation of cracking were measurable and different for modified and unmodified binders. An unmodified asphalt section lost only between 25% and 33% of the original modulus before the initiation of surface cracks, whereas a polymer-modified asphalt lost between 62% and 57%. The cracking was top-down given the aged condition of the test sections.

Lessons in bridge damage learned from the Wenchuan earthquake

A strong earthquake occurred in Wenchuan County, Sichuan Province, China, on May 12, 2008. Shortly after the earthquake, the Turner-Fairbank Highway Research Center of the Federal Highway Administration, in partnership with the Research Institute of Highways, the Ministry of Communication of China, led a reconnaissance team to conduct a post-earthquake bridge performance investigation of the transportation system in the earthquake affected areas. The U.S. transportation system reconnaissance team visited the area during July 20–24, 2008. This paper presents the findings and lessons learned by the team.

Two full-scale loading experiments of geosynthetic-reinforced soil (GRS) abutment wall

Two full-scale segmental facing Geosynthetic-Reinforced Soil (GRS) abutment walls, referred to as "the NCHRP test abutments", were constructed and load-tested at the Turner-Fairbank Highway Research Center in McLean, Virginia. The two experiments were conducted to (1) examine the behavior of segmental facing GRS abutment walls under increasing vertical loads on a bridge sill, and (2) furnish a set of well-defined data for verification of analytical models. This paper describes the two experiments and the measured and observed behavior due to the load applications. The behavior of the test walls were discussed and compared with existing performance criteria that were developed based on experience with real bridges. The comparisons were made in terms of maximum sill settlement, maximum angular distortion, and maximum lateral wall movement. In addition, the safety factors and failure loads of the test walls were evaluated by the prevailing design method. The predicted failure loads obtained from the prevailing design method, even with all safety factors and strength reduction factors being set equal to one, were found to be much lower than the measured values, suggesting that the prevailing design method is likely to be overly conservative.

Nonrecovered compliance from dynamic oscillatory test vis-à-vis nonrecovered compliance from multiple stress creep recovery test in the dynamic shear rheometer†The opinions, findings and conclusions expressed in this document are those of the author and not necessarily of the Federal Highway Administration or other researchers at the Turner-Fairbank Highway Research Center.

Nonrecovered compliance is presently being viewed as the most appropriate rheological parameter for evaluating the propensity of an asphalt binder to resist permanent deformation or rutting in the pavement wheel paths. The nonrecovered compliance can be obtained in two ways using two different types of tests in the dynamic shear rheometer. One is through the dynamic oscillatory test using a frequency, time, strain or stress sweep, wherein the data generated is in terms of the complex modulus and phase angle that can be used in a proper mathematical form to obtain the nonrecovered compliance. The other is through the multiple stress creep and recovery test wherein the nonrecovered strain at each stress level after 10 cycles of creep and recovery is divided by the stress value to obtain the nonrecovered compliance. The two methods are very different in the way data gets generated and handled and, also, in the level of ease or difficulty in data generation. With two methods competing to come up with the same outcome, it is important to see a comparison of the merits and demerits of each of the methods in order to ascertain which of the two is more reliable, more fundamental, easier to use and better to implement. The purpose of this paper is to provide a one-on-one comparison between the two methods using the same set of binders and also to look at the possible benefits of one over the other.

An assessment of modeling strategies for composite curved steel I-girder bridges

This paper investigates the qualities and limitations of a number of modeling strategies for design analysis of composite curved steel I-girder bridges. The strategies range from modified line-girder analysis to 3D finite element approaches. Several key modeling considerations, including girder web distortion, the idealization of the cross-frames, support and load height, and displacement compatibility between the girders and slab, are addressed. Approximate approaches are investigated for the capture of composite I-girder web-distortion effects when modeling the girders using open-section thin-walled beam theory. A representative full-scale composite curved I-girder bridge, tested at the Federal Highway Administration (FHWA) Turner-Fairbank Highway Research Center (TFHRC), is utilized for assessment of the different approaches. The predictions of key lateral and vertical displacements, reactions, cross-frame forces, major-axis bending stresses and flange lateral bending stresses are evaluated using the test bridge. The paper highlights a number of subtle but critical blunders that can occur in the context of the different methods. Engineers need to be aware of these pitfalls in order to avoid potential problems due to inaccurate analysis predictions. The paper closes with a discussion of the efficacy of the various approaches.

Effect of Coefficient of Thermal Expansion Test Variability on Concrete Pavement Performance as Predicted by Mechanistic-Empirical Pavement Design Guide

The coefficient of thermal expansion (CTE) of concrete is a property that can affect the performance of the pavement and its service life and is one of the most important inputs in the Mechanistic-Empirical Pavement Design Guide (MEPDG). The CTE can be either estimated or measured in the laboratory. The test method used to determine this property is AASHTO TP 60, still a provisional test method and not yet evaluated for its precision. CTEs of more than 1,800 concrete specimens were measured at the Turner-Fairbank Highway Research Center. The specimens included cylinders that were cast in the laboratory as well as field cores obtained from the Long-Term Pavement Performance pavement sections. Approximately 150 of the specimens were tested individually several times for assessment of repeatability of the test method. An analysis is presented of test differences observed, as is a sensitivity analysis of the CTE test variability on predicted performance based on the MEPDG. The differences in predicted international roughness index (IRI), percent slabs cracked, and faulting due to test variability were determined for concretes with CTEs ranging from 4 to 7 × 10−6 in./in./°F. It was observed that differences in test results may result in significant discrepancies in the predicted IRI, percent slabs cracked, and faulting. Thus, a single test result should not be used as representative of the CTE of a mixture due to the considerable impact of the test variability on the predicted pavement performance. Moreover, the specifications should state the minimum number of tests necessary for the CTE determination and the acceptable test variability.

Bulb-T Beams with Self-Consolidating Concrete on Route 33 in Virginia

A study investigated bulb-T beams with self-consolidating concrete (SCC) in the Route 33 Bridge over the Pamunkey River in Virginia, near West Point. Before construction of the bridge, two test beams with SCC similar in cross section to the actual beams in the structure were cast and loaded to failure at FHWA's Turner-Fairbank Highway Research Center in McLean, Virginia. They were tested for transfer length, development length, flexural strength, and shear strength. These test beams demonstrated that SCC members can be designed with the same methods, assumptions, and limiting values as used for normally consolidated concrete beams. On the basis of positive results, beams with SCC were cast and placed in the bridge. It was found that SCC can be prepared yielding adequate slump flow without segregation, satisfactory strength and acceptably low permeability. Attention should be paid to the sensitivity of SCC to water content.

Analysis of Surface Inertial Profiles Measured on Jointed Portland Cement Pavements

Abstract The feasibility of measuring slab curling using high-speed profilers was assessed by the Turner-Fairbank Highway Research Center of the Federal Highway Administration. The primary objectives of the study were to: (1) measure the surface profile of the concrete slabs using state-of-the-art profiling technology, (2) establish a standard procedure to identify warping and curling of the slabs using the measured surface profiles, (3) relate the changes in pavement surface profiles to the change in surface temperature, (4) determine the relationship between changes in pavement surface profile, roughness, premature transverse cracking, and construction condition, and (5) establish a standard procedure to analyze the condition of joints, including rotation and faulting, and determine the relationship between joint condition, and roughness. The paper discusses data mining on a database of 16 000 slabs measured during four time windows in a 24-h period. A visualization of all parameters and samples at all four times of the day show the groups of slabs which have cracked and are likely to crack shown by the divergence from the normal response measured in most slabs.

Determining Colors for Traffic Control Devices at Transponder-Controlled Tollbooth Lanes with a Sign Simulator

The number of toll roads is increasing in the United States, and drivers face signing practices that vary greatly across jurisdictions. In addition to the increase in toll roads, there is an increase in the use of electronic toll collection (ETC) transponders. The present study was undertaken to develop standard signing practices—specifically for background color, legend color, underlay color (i.e., the background color for the pictographs), and pictograph—for ETC-supported toll road signs. The study reviewed selected ETC toll road signs in the United States to help select candidate elements and options to be used in a laboratory experiment. On the basis of that review, 120 signs depicted on 35-mm slides were developed with a standard sign-design software package. The signs were presented to 60 participants (equally divided by sex and age group) in the recently updated Turner–Fairbank Highway Research Center's sign simulator. The signs were shown to participants starting at a simulated distance of approximately 1,125 ft (343 m), and then the image was zoomed toward the participants at a constant speed controlled by the laboratory's computer. Each participant's detection distance, guidance legibility distance, and pictograph legibility distance were recorded. The results showed that, overall, green as a background color obtained the longest guidance information legibility distance. Also, fonts that provided the highest contrast to the background color (such as white on black) were the most effective for legibility. Underlay colors that showed the highest contrast to the pictographs were the most effective and included all of the lighter colors tested (white, yellow, and blue).

Determining Colors for Traffic Control Devices at Transponder-Controlled Tollbooth Lanes with a Sign Simulator

The number of toll roads is increasing in the United States, and drivers face signing practices that vary greatly across jurisdictions. In addition to the increase in toll roads, there is an increase in the use of electronic toll collection (ETC) transponders. The present study was undertaken to develop standard signing practices—specifically for background color, legend color, underlay color (i.e., the background color for the pictographs), and pictograph—for ETC-supported toll road signs. The study reviewed selected ETC toll road signs in the United States to help select candidate elements and options to be used in a laboratory experiment. On the basis of that review, 120 signs depicted on 35-mm slides were developed with a standard sign-design software package. The signs were presented to 60 participants (equally divided by sex and age group) in the recently updated Turner-Fairbank Highway Research Center's sign simulator. The signs were shown to participants starting at a simulated distance of approxi...

Virginia Department of Transportation Early Experience with Self-Consolidating Concrete

This paper summarizes the work conducted by the Virginia Department of Transportation on self-consolidating concrete (SCC). Initially, a laboratory study with locally available materials was conducted to develop and evaluate the properties of SCC, including flow, segregation, strength, permeability, resistance to cycles of freezing and thawing, and drying shrinkage. Studies at two precast plants followed. Then, one of the precast plants furnished SCC for an arch bridge. The other plant fabricated two bulb T-beams with SCC. Studies also were conducted with cast-inplace SCC. A structures pad at the FHWA Turner—Fairbank Highway Research Center, a median on I-64, and a column in Norfolk, Virginia, were constructed with SCC. The results show that SCC can be produced successfully and provide many benefits to transportation agencies and the construction industry.

Measurement and Significance of the Coefficient of Thermal Expansion of Concrete in Rigid Pavement Design

The coefficient of thermal expansion (CTE) is a fundamental property of concrete. It has long been known to have an effect on joint opening and closing in jointed plain concrete pavement, crack formation and opening and closing in continuously reinforced concrete pavement, and curling stresses and thermal deformations in both types of pavements. However, it has not been included as a variable either in materials specifications or in the structural design of concrete pavements. Hundreds of cores were taken from Long-Term Pavement Performance sections throughout the United States and were tested by FHWA's Turner–Fairbank Highway Research Center laboratory, using the AASHTO TP 60 test procedure. The CTE values were then assimilated into groups on the basis of aggregate types, and the mean and range of CTE were calculated. These results were then used in the new mechanistic–empirical pavement design guide to determine the significance of the measured range of CTE on concrete pavement performance. The CTE of the concrete was found to vary widely, depending on the predominant aggregate type used in the concrete. Sensitivity analysis showed CTE to have a significant effect on slab cracking and, to a lesser degree, on joint faulting. Its overall effect on smoothness was also significant. Given that CTE has not been used before in routine pavement structural design, the conclusion is that this design input is too sensitive to be ignored and must be fully considered in specifications and in the design process to reduce the risk of excessive cracking, faulting, and loss of smoothness.

Virginia Department of Transportation Early Experience with Self-Consolidating Concrete

This paper summarizes the work conducted by the Virginia Department of Transportation on self-consolidating concrete (SCC). Initially, a laboratory study with locally available materials was conducted to develop and evaluate the properties of SCC, including flow, segregation, strength, permeability, resistance to cycles of freezing and thawing, and drying shrinkage. Studies at two precast plants followed. Then, one of the precast plants furnished SCC for an arch bridge. The other plant fabricated two bulb T-beams with SCC. Studies also were conducted with cast-in-place SCC. A structures pad at the FHWA Turner-Fairbank Highway Research Center, a median on I-64, and a column in Norfolk, Virginia, were constructed with SCC. The results show that SCC can be produced successfully and provide many benefits to transportation agencies and the construction industry.

Measurement and Significance of the Coefficient of Thermal Expansion of Concrete in Rigid Pavement Design

The coefficient of thermal expansion (CTE) is a fundamental property of concrete. It has long been known to have an effect on joint opening and closing in jointed plain concrete pavement, crack formation and opening and closing in continuously reinforced concrete pavement, and curling stresses and thermal deformations in both types of pavements. However, it has not been included as a variable either in materials specifications or in the structural design of concrete pavements. Hundreds of cores were taken from Long-Term Pavement Performance sections throughout the United States and were tested by FHWA's Turner-Fairbank Highway Research Center laboratory, using the AASHTO TP 60 test procedure. The CTE values were then assimilated into groups on the basis of aggregate types, and the mean and range of CTE were calculated. These results were then used in the new mechanistic-empirical pavement design guide to determine the significance of the measured range of CTE on concrete pavement performance. The CTE of t...

Applications of Laser-Based Instrumentation for Highway Bridges

This paper describes a Federal Highway Administration Nondestructive Evaluation Validation Center (NDEVC) project to develop tools to better measure large-scale structures. The specific instrument discussed is a laser-based instrument that has been successfully used in a number of applications related to highway bridges. To help understand the way the instrument is used, a description of the basic system operation is given, followed by a detailed explanation of the main instrument features. Several projects conducted with this laser system are then presented as examples of the capabilities of the instrumentation. Although the system does have limitations, especially in terms of measurement range and cost, the system has proven to be versatile for a wide range of applications. Both laboratory and field applications are given. Most of the laboratory tests with the instrument have occurred at the NDEVC, which is located at the Turner-Fairbank Highway Research Center in McLean. Virginia. Field tests have taken place at several locations around the United States.

Analytical and experimental study of a skew bridge model

This paper describes an experimental study of a skew bridge model conducted at the Federal Highway Administration Turner-Fairbank Highway Research Center. The objectives of the experiment are: (1) to perform a pilot study on the design, construction, instrumentation, testing and data processing of a skew bridge model, (2) to provide experimental data to validate a 3D finite element model developed for the skew bridge, and (3) to confirm the applicability of a simplified bridge model (the dual-beam stick model) developed for the dynamic analysis of skew bridges. Details of the design, construction, instrumentation, testing and data processing of the model bridge are described. Results for static displacements, natural frequencies, mode shapes and damping of the model bridge will be presented. A comparison of the results obtained from these tests with those obtained numerically from a finite element analysis and analytically from the dual stick bridge model shows that good correlation is obtained.