This session covers the topics of pavement surface characteristics: friction, texture, and smoothness.
Friction and Surface Texture Evaluation of Green-Colored Bike Lanes
Bouzid Choubane, Florida Department of TransportationShow Abstract
Edward Offei, Applied Research Associates, Inc. (ARA)
Guangming Wang, Florida Department of Transportation
Charles Holzschuher, Florida Department of Transportation
Interest in colored treatment on bicycle lanes and crossings has been growing in the United States in recent years. In comparison, this practice, has been prevalent in European cities for longer time. It was not until 2011 that the green colored treatment received official interim approval from the Federal Highway Administration (FHWA) for experimental use on bicycle facilities across the country (1). This study focused primarily on evaluating the friction and texture characteristics of five (5) independent green colored bicycle lane projects consisting of either (1) Epoxy Modified, (2) Thermoplastic, or (3) High Friction Surface Treatment materials in Florida. A total of three types of existing pavement surfaces (concrete, open and dense graded asphalt pavements) were used as substrate for the colored application. These chosen sites include both control test sections representing the bike lanes with limited/no traffic interaction and keyhole sections that represent traffic conflict areas (areas where bicycles and vehicles come into conflict). The friction and texture values were obtained using the Dynamic Friction Tester (DFT) and Circular Texture Meter (CTM), respectively. Results indicated that all green bike lane projects met the initial friction number requirements for Florida’s Patterned Textured Pavements. Minor friction loss was observed at the keyhole sections when compared to the control sections indicative of traffic wear effects. Factorial Analysis of Variance (ANOVA) showed that factors such as pavement surface type as well as type of green bike lane material applied and the presence of traffic wear have significant influence on the friction values. In addition, based on mean profile depth (MPD) measurements, only the interaction of pavement surface type and the bike lane treatment type had significant impact on the texture. The presence of traffic was not a significant factor. All these results ultimately lead to new design criteria in 2016 permitting a more wide-spread application of green colored bike lanes on the Florida State Highway System.
Connected-Vehicle Method of Estimating International Roughness Index
Raj Bridgelall, North Dakota State UniversityShow Abstract
Md Tahmidur Rahman, Fugro USA Land Inc.
Denver Tolliver, Upper Great Plains Transportation Institute
Jerry Daleiden, ARRB Group Inc.
The high cost of deploying inertial profiler vehicles to evaluate pavement roughness limits data collection to relatively small portions of the network. Previous research demonstrated a connected vehicle method that enables continuous roughness evaluations for all roadways, including local and unpaved roads. This research estimates the international roughness index from the connected vehicle method that produces a roughness index called the road impact factor. The authors establish a theory that the ratio between the road impact factor and the international roughness index changes with the dominant profile wavelengths. Case studies validated the theory by evaluating the ratio of the two roughness indices across five different pavement types, using the same vehicle at a fixed speed. The ratios distributed normally. The spread of the distribution reflected expected differences in the dominant spatial wavelength among the different surfaces. The normal distribution anticipates that the precision of estimating previously unknown international roughness indices from the measured road impact factors will continuously increase with traversal volume.
Prediction of Asphalt Mixture Surface Texture Level and Its Distributions Using Mixture Design Parameters
DE CHEN, Southwest Jiaotong UniversityShow Abstract
Sen Han, Chang'an University
Cheng Ling, Pike Industries, Inc.
Qian Su, Southwest Jiaotong University
Pavement skid resistance plays a key role in traffic safety. Meanwhile, tire-pavement noise is a major source of the traffic noise in urban areas. Current asphalt mixture design methods, however, mostly focus on volumetric and mechanical properties and pay little attention to the skid resistance and noise reduction performance of asphalt mixtures, which are significantly affected by the surface textures of the asphalt mixtures. Incorporating the evaluation of surface texture into the mixture design would aid in a more rational selection of materials considering both mechanical and functional properties of asphalt mixtures. In this paper, the surface texture properties of several types of asphalt mixtures are measured using a recently developed 2-Dimensional Image Texture Analysis Method (2D-ITAM). A prediction model correlating the mixture surface texture levels at different central texture wavelength in octave band with the important mix design parameters is established using a multivariate non-linear regression analysis. The model is validated through the laboratory test and imaging measurement indicating its capability of predicting the level and distributions of mixture surface texture. The prediction model is anticipated to provide a basis of optimized mixture design considering the skid resistance and noise reduction performances of asphalt pavement.
Highway Drainage at Superelevation Transitions by 3-D Computational Fluid Dynamics Modeling
Lu Chen, Virginia Polytechnic Institute and State UniversityShow Abstract
Francine Battaglia, Virginia Polytechnic Institute and State University
Gerardo Flintsch, Virginia Polytechnic Institute and State University
David Kibler, Virginia Polytechnic Institute and State University
This paper focuses on the development of the three-dimensional (3D) computational fluid dynamics (CFD) model for predicting and understanding the hydrodynamics of water accumulation and drainage on multi-lane highway at superelevated transitions. The model was refined and used to simulate different road geometries and conditions based on real road coordinates. Fluid quantities have been solved based on the conservation of mass and momentum in a 3D domain to capture the detail of pavement conditions as well as the physics of the fluids flow. The model was first validated based on highway straight sections with no cross slope against previous one-dimensional (1D) kinematic wave (KW) model and then extended to superelevation transitions. The sensitivity analysis on superelevation transitions includes the impacts on water film thickness (WFT) distribution due to different factors, such as longitudinal slope, cross slope, number of lanes and rainfall rate. Further comparisons of the real road results between 1D and 3D models have shown that 1D model over-predicts WFT and is not sensitive to the number of lanes. This is attributed to the fact that the cross slope is represented only through the calculation of total drainage path length and is not included directly in the 1D KW runoff calculation under steady rainfall conditions.
Ride Comfort Analysis of Passenger Cars on Mountain Expressways Based on Triaxial Acceleration Data from Field Driving: Case Study in China
Jin Xu, Chongqing Jiaotong UniversityShow Abstract
Kui YANG, Southwest Jiaotong University
Wei Lin, Chongqing Jiaotong University
Yiming Shao, Chongqing Jiaotong University
To acquire the ride comfort level and its key influencing factors on mountain expressways, real vehicle tests under natural driving behavior were conducted to collect the acceleration data from a variety of small passenger cars. From these data, the peak cumulative frequency curves, characteristic percentiles, and root-mean-square (RMS) values of tri-axial acceleration were obtained. The results showed that the magnitude and characteristic percentiles of acceleration rate were slightly lower than those of the deceleration rate. Different roads and car models varied in the longitudinal acceleration/deceleration but showed an overall high level of longitudinal comfort as indicated by most data being below the comfort threshold. The measured lateral acceleration data varied significantly among different car models and different roads. Overall, the mountain expressways provided good lateral comfort. In terms of the vertical direction, the passengers only rarely experienced discomfort in the lightest vehicle in very few locations; furthermore, vibration marking and colored strips on the pavement did not cause discomfort for occupants in passenger cars. On the expressways, the differences in acceleration amplitude and RMS value between three axial directions were negligible. The RMS value of acceleration in each axial direction and the weighted RMS values in the three axial directions ranged from “comfortable” to “slightly uncomfortable” levels. The evaluation results were consistent with the actual riding experience of the drivers and passengers.
Use of Smartphone to Measure Pavement Roughness Across Multiple Vehicle Types at Different Speeds
John StriblingShow Abstract
William Buttlar, University of Missouri, Columbia
Shahidul Islam, Applied Research Associates, Inc. (ARA)
William Vavrik, Applied Research Associates, Inc. (ARA)
Measuring pavement roughness has been common practice for transportation agencies for many years now. Normally, this measurement is recorded as International Roughness Index (IRI) and is collected via a data collection vehicle (DCV) using lasers, accelerometers, and distance measuring devices. Unfortunately, this practice is cost-prohibitive to conduct on an annual basis. However, recent studies have examined the use of low-cost equipment, such as smartphones, to capture the same data and then analyze that data with in-house software. One of the primary goals with this new method is to engineer the capability to crowdsource roadway roughness data collection. This would allow asset managers to have current, and potentially real-time, roughness data from which strategic decisions can be made. The challenge with this transition is to correlate roughness data collected from different sized vehicles, going different speeds, and operating in different environments to the standard “golden car” model as used in standard IRI measurements. This study builds off of previous work whereby a smartphone app was used to collect roughness data and then analyzed via an in-house MATLAB program to produce accurate IRI measurements. That same smartphone app and in-house software was used to analyze data collected from different sized vehicles travelling at different speeds. These measurements were then compared to an official IRI measurement collected just a few months prior. The results demonstrate clear sensitivity to those parameters, thereby opening the door for calibration of the software to account for these variables arising from data crowd sourcing.
Road Surface Texture Evaluation Using 3-D Laser Data
Humaira Zahir, Kansas State UniversityShow Abstract
Mustaque Hossain, Kansas State University
Shuvo Islam, Kansas State University
Richard Miller, Kansas Department of Transportation
Road surface friction helps minimize skidding and reduces the number of roadway crashes. Because skid resistance depends on pavement texture, quantification of texture characteristics provides specific information about the friction condition of the roadway. In order to assess how surface texture varies with pavement surface type, roadways with different surface types were evaluated. Data was collected using the Laser Crack Measurement System (LCMS) three-dimensional laser profiler and ASTM Locked Wheel Skid Trailer to determine the correlation between texture depth and skid number. This study objective was to find a suitable correlation between skid number and texture depth so that the traditional skid trailer can be supplemented by LCMS for routine skid monitoring. Results show that a good correlation exists between skid number and texture depth in the range of 0.5 to 1.5 mm of texture depth for all three surface types studied.
Accuracy of Road Surface Profilers in TRUE Project: Experiment to Compare Test Methods for Surface Roughness Under Actual Road Environment
Kazuya Tomiyama, Kitami Institute of TechnologyShow Abstract
Hiroyasu Nakamura, NIPPO Corporation
Hiroyuki Mashito, Toa Road Corporation
Masakazu Jomoto, Taisei Rotec Corporation
Kazuhiro Watanabe, Public Works Research Institute
A number of surface roughness measuring devices has been introduced in Japan since the Road Bureau of Ministry of Land, Infrastructure, Transport and Tourism has issued a strategy for inspection of the whole road stock and has provided an implementation guide for evaluating roughness by means of the International Roughness Index (IRI) in 2013. Against this background, the specified nonprofit organization Pavement Diagnosis Researchers Group (PDRG) in Japan conducted the experiment to compare test methods for surface roughness under actual road environment (TRUE Project). Thirty-four devices including high- and low- speed profilers and response type systems participated in the first experiment of the project held at Hokkaido, Japan in 2014. This paper analyzes the accuracy of the high- and low speed profilers in terms of repeatability and reproducibility as well as portability for both IRI and profile measurements. The influence of operating speed on the measurements is also considered for the high-speed profilers. The final purpose of this study is to provide the essential information for associating different profilers so that profiler users and developers can assess the accuracy of each profiler to produce standard measures of roughness to interchange inspection results. For this purpose, we demonstrate a tolerance of the repeatability, reproducibility, and portability and the influence of operating speed for the high-speed profilers in respect to the IRI and profile measurements obtained in the experiment. The information described in this paper is of benefit to profiler developers as well as users for development and improvement of their profilers.
Viscoelastic Model for Estimating the International Roughness Index by Smartphone Sensors
Chih-Sheng Chen, National Taiwan UniversityShow Abstract
Chiapei Chou, National Taiwan University
Ai-Chin Chen, National Taiwan University
This study developed a viscoelastic model (VM) that uses smartphone acceleration data to estimate international roughness index (IRI). The developed VM is simpler and more straightforward than the majority of previous models that have a similar function. The most notable benefit of the VM is that vehicle suspension parameters need not be measured or known. Fourteen profile samples from ProVAL 3.5 were selected for VM validation. The estimated IRIs from the VM showed strong correlations with the IRIs calculated by ProVAL. The proposed model was further verified by comparing the analyzed results of 39 field test sections with the calculated IRIs from ProVAL under two cases, namely, agency application and crowdsourcing application. In the first case, the suspension parameters were calibrated by least square method using the available field inertial profiler data. In the second case, golden-car parameters were used in the developed VM to simulate the crowdsourcing application. Both cases indicate that the VM is a promising tool for estimating the IRI. The IRI linear correlation between the model outputs of these two cases and the ProVAL calculation are R2= 0.91 and 0.89, respectively.
Effect of Road Roughness and Vehicle Speed on Dynamic Load Prediction and Pavement Performance Reduction
Boris Goenaga, Universidad del NorteShow Abstract
Luis Fuentes, Fundacion Universidad del Norte
Otto Mora, Universidad del Norte
Traffic constitutes a fundamental parameter in the analysis, design and performance prediction of pavement structures. Although the current Mechanistic Empirical Pavement Design Guide (MEPDG) uses axle load spectra to characterize the traffic variable for pavement design purposes, pavements around the world continue to be designed using the Equivalency Single Axle Load (ESAL) concept, which is based on the static load of the vehicles (dead weight). However, the dynamic loads induced by roughness can be considerably higher than the static load in specific locations of a pavement section, causing an unexpected adverse impact on the performance of pavement structures. In the present investigation, the effects of both pavement roughness and vehicle speed on the dynamic loads developed at the tire pavement interface are evaluated along with their effects on the performance of pavement structures. In other to achieve the above objective, 787 pavement profiles were analyzed, combining rural and urban environments, as well as rigid and flexible pavement sections. The dynamic load produced at the tire pavement interface for all pavement profiles was modeled. Two roughness indices, The International Roughness Index (IRI) and the Dynamic Load Index (DLI), were determined for each profile. A correlation model between the IRI and the DLI was developed. Additionally, a Traffic Correction Factor (TCF) was proposed to account for the dynamic load effects induced by roughness and vehicle speed. The proposed TCF could be used to modify the estimation of traffic damage on road sections with high roughness levels, therefore improve future performance prediction processes. Finally, a methodology to calculate the reduction of the remaining life of a pavement structure due to the surface roughness was proposed.
Investigation of Slab Curvature in LTPP SPS-2 Experiment Using Empirical Mode Decomposition of Pavement Profilometer Data
Derek Tompkins, American Engineering Testing, Inc.Show Abstract
Dan Franta, Minnesota Department of Transportation
Kyle Hoegh, Minnesota Department of Transportation
Lev Khazanovich, University of Pittsburgh
The concept of built-in curl is uniformly accepted in the research community as an important characteristic of the pavement structure. To this end, the AASHTO M-E procedure considers built-in curl as a parameter in its rigid design, and the procedure is very sensitive to this parameter. Prior research has used slab curvature to estimate the built-in curl parameter. The authors follow on this research in the development of analytical tools to infer slab profile directly from profilometer data. The developed tools represent an automated empirical mode decomposition (EMD) process contained within the Hilbert-Huang Transform (HHT). These tools were then applied to profilometer data from the Long-Term Pavement Performance (LTPP) database. It was found that slab profile could be recovered for LTPP projects; for these projects, the slab profiles could be correlated with built-in curl over the pavement life. The application of these tools presents a promising first step in the extension of profilometer data beyond performance measures (i.e., smoothness) and into areas that directly influence the mechanistic-empirical modelling of the pavement structure.
Investigating Predictability of Pavement Friction on Rural Roads in Ontario, Canada
Luciana Omar, Carleton UniversityShow Abstract
Abd El Halim, Carleton University
Karim Ismail, Carleton University
Pavement friction is one of the most important operational requirements to ensure highway safety. Pavement friction contributes to the driver’s ability to maintain vehicle control during braking and cornering maneuvers. Pavement surface friction is affected by pavement surface characteristics, vehicle operational parameters, tire properties, and environmental conditions. Among these factors, transportation agencies are responsible for monitoring and maintaining adequate pavement surface texture which is affected by two level of textures, microtexture and macrotexture. Practically, incorporating friction into the pavement management is still challenging for many transportation agencies because there is not a specific device that can measure both textures simultaneously. In Ontario Canada, microtexture is measured using a locked wheel device and macrotexture is measured using laser profilometer attached to high-speed road analyzer (ARAN). The former test is relatively time-consuming and often done with less frequency than macrotexture. This research aimed to examine variables that affect microtexture and macrotexture and to develop a regression model that can be used by the road agencies to estimate skid resistance using mixture parameters, traffic characteristics, and direct measurement of macrotexture using ARAN. In this study, field data was collected on freeway segments of asphalt surface course within a variety of functional classes of rural roads of the Ontario road network. Stepwise regression approach was conducted to identify significant variables and build the prediction model. The results show that pavement skid resistance for asphalt surfaces can be estimated with a fair coefficient of determination
Mean Texture Depth and Deflection Detection and Steering Angle-Based Bike Pavement Condition Assessment
Xiaodong Qian, University of California, DavisShow Abstract
Deb Niemeier, University of California, Davis
Davis’ reputation as the “Bicycle Capital of the U.S.” is due to its high number of cyclists, its renowned system of bikeways and cyclist-friendly facilities, and supportive City and University programs. But much of the infrastructure is showing wear, with pavement surface defects such as cracks, uplifting, and pothole. To date, bicycle paths or lanes have not emerged as key priorities in traditional pavement systems analysis. Most cities rely on route preferences (e.g. a common route to school) or visual checks to prioritize pavement conditions for bicycles. This research investigates pavement smoothness, bicycle displacement and steering and their relationship to ride quality. Specifically, we examine the relationship between bicycle ride quality and traditional pavement roughness measurement, and a new deflection displacement and steering angle indices. We used 57 bike path sections with a representative range of pavement surface conditions to collect acceleration data, steering angle data, GPS location data, and mean texture depth (MTD) data. We also recruited cyclists to complete a post-ride survey on ride quality. As has been noted in the, albeit sparse literature, we found that bicyclists’ comfort is greatly influenced by surface roughness. We also found that vibration in vertical direction plays an important role in ride quality, regardless of bicycle type. Our results can help to shape a new bike pavement condition assessment system using MTD, steering angle, and deflection displacement.