Evaluation of the Repeatability and Reproducibility of Network-Level Pavement Macrotexture Measuring Devices Jacquelyn Nicole Keeney Thesis submitted to the faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements for the degree of Masters of Science In Civil Engineering Gerardo W. Flintsch, Chair Saied Taheri Samer W. Katicha 7-6-2017 Blacksburg, VA Keywords: Macrotexture, Network-Level, MPD, Repeatability, Reproducibility Evaluation of the Repeatability and Reproducibility of Network-Level Pavement Macrotexture Measuring Devices Jacquelyn Keeney ABSTRACT The purpose of this thesis was to assess the repeatability and reproducibility of two high-speed macrotexture measuring systems. The first portion of the study collect ed macrotexture measurements using the two high- speed systems on the Virginia Smart Road facility and validated the reproducibility of the mean profile depth (MPD) measurements with reference CT Meter measurements. The various data sets were then compared with each other. The objective was to determine whether the two systems are collecting repeatable and reproducible data. The analysis showed that the two high-speed systems investigated have good repeatability (approximately 0.11 mm) when measuring the average MPD of the sections investigated. The two systems produce measurements that are highly correlated with the reference ones obtained with the CT Meter. While the Ames systems, with the data processed using the Virginia Tech filter, measures MPD values that are very close to those of the CT Meter, with a virtually zero systematic bias, the SCRIM measure slightly lower readings. The differences are thought to be due to the filtering of the data used by the SCRIM processing software to eliminate dropout and spikes in the laser measurements Evaluation of the Repeatability and Reproducibility of Network-Level Pavement Macrotexture Measuring Devices Jacquelyn Keeney GENERAL AUDIENCE ABSTRACT The researched compared two different devices used to record data for a surface property of pavement, macrotexture. The macrotexture is a measure of the spaces in between the particles making up the surface of the asphalt. This property is linked to the level of friction on the roadways affecting safety. The readings from each of the devices were validated through a reference device in order to insure they were obtaining reliable results on the Virginia Smart Road. The repeatability and reproducibility of each of the devices was examined to determine whether the two systems were collecting repeatable and reproducible data. It was determined the two devices had good repeatability and were highly correlated with the reference device. The first device called the Ames system obtained measurements very close to those of the reference device, with a virtually zero systematic bias. The second device called the SCRIM system obtained measurements close to the reference device but slightly lower. Through the comparisons and data analysis, an equation permitting users to convert readings from the SCRIM to Ames system and vice versa was computed. Because one device obtains slightly higher readings than the other, this equation is very important in order to have comparable results. CONTENTS Abbreviations ............................................................................................................................................... vi List of Tables ................................ ................................ ................................ ................................ .............. vii List of Figure s ................................ ................................ ................................ ................................ ............ viii Chapter 1: Introduction ................................ ................................ ................................ ................................ . 1 Problem Statement ................................ ................................ ................................ ................................ .... 2 Objective ................................ ................................ ................................ ................................ ................... 2 Significance/ Contribution ................................ ................................ ................................ ........................ 2 Overview ................................ ................................ ................................ ................................ ................... 3 Chapter 2: Literature Review ................................ ................................ ................................ ........................ 4 Uses of Macrotexture ................................ ................................ ................................ ................................ 4 Macrotexture Measurement ................................ ................................ ................................ ...................... 4 Macrotexture Characterization Parameters ................................ ................................ ........................... 4 “Static” Macrotexture Measuring Devices ................................ ................................ ............................ 6 High-Speed Macrotexture Measuring Devices ................................ ................................ ..................... 6 High-Speed Macrotexture Devices Used in This Thesis ................................ ................................ .......... 7 Data Processing ................................ ................................ ................................ ................................ ......... 8 Past Equipment Comparisons ................................ ................................ ................................ ................. 10 Conclusions of the Literature Review ................................ ................................ ................................ ..... 12 Chapter 3: Research Approach ................................ ................................ ................................ ................... 13 Data Collection and Pre-Processing ................................ ................................ ................................ ........ 13 Data Analysis ................................ ................................ ................................ ................................ .......... 15 Descriptive Statistics ................................ ................................ ................................ ............................... 15 Limits of Agreement Analysis ................................ ................................ ................................ ................ 16 Orthogonal Regression Analysis ................................ ................................ ................................ ............. 18 Chapter 4: Results ................................ ................................ ................................ ................................ ....... 19 Reference Measurements ................................ ................................ ................................ ........................ 19 Qualitative Comparisons ................................ ................................ ................................ ......................... 20 Repeatability ................................ ................................ ................................ ................................ ........... 22 Reproducibility ................................ ................................ ................................ ................................ ....... 22 Orthogonal Regression Analysis ................................ ................................ ................................ ............. 28 Discussion ................................ ................................ ................................ ................................ ............... 31 Chapter 5: Summary, Findings, and Conclusions ................................ ................................ ....................... 32 Summary ................................ ................................ ................................ ................................ ................. 32 iv Main Findings of the Research ............................................................................................................... 32 Conclusions ............................................................................................................................................. 32 Recommendations ................................................................................................................................... 33 References ................................................................................................................................................... 34 v ABBREVIATIONS MTD – Mean Texture Depth PMS – Pavement Management Systems MPD – Macrotexture Mean Profile Depth CT Meter – Circular Track Meter SCRIM – Sideways-Force Coefficient Routine Investigation Machine ASCE - American Society of Civil Engineers SMTD - Sensor-Measured Texture Depth FHWA - Federal Highway Administration DOT – Department of Transportation TDOT - Texas Department of Transportation GGD - General Gaussian Distributions FDR - False Discovery Rate LOA – Limits of Agreement Anova – Analysis of Variance MS – Mean Square vi LIST OF TABLES Table 1. Lengths of Virginia Smart Road sections recorded in meters....................................................... 15 Table 2. CT Meter MPD means and standard deviations for examined pavement sections ....................... 19 Table 3. High-Speed Macrotexture measurements ..................................................................................... 20 Table 4. ANOVA test results ...................................................................................................................... 22 Table 5. Repeatability (mm) for the average sectional MPD ...................................................................... 22 Table 6. Limit of Agreement Boundary Results: Concrete 4 Removed ..................................................... 27 Table 7. Ames and CT Meter orthogonal regression analysis results (Concrete 4 removed) ..................... 28 Table 8. SCRIM and CT Meter orthogonal regression analysis results (Concrete 4 removed) .................. 29 Table 9. SCRIM and Ames orthogonal regression analysis results: Concrete 4 removed .......................... 30 vii LIST OF FIGURES Figure 1. Method for computing the mean segment depth (Pulugurtha et al 2012) ..................................... 5 Figure 2. The CT Meter ................................................................................................................................ 6 Figure 3. The SCRIM system ....................................................................................................................... 7 Figure 4. The Ames system .......................................................................................................................... 8 Figure 5. Test site layout ............................................................................................................................. 14 Figure 6. Comparison of MPD profiles ...................................................................................................... 17 Figure 7. Standard deviation of the CT Meter MPD measurements vs. the CT Meter MPD means .......... 19 (a) Figure 8. Comparison of Ames and SCRIM 1-meter data (b) 21 Figure 9. Standard deviation vs. mean for the high-speed systems: (a) Ames, (b) SCRIM ....................... 21 Figure 10. Ames vs. CT Meter comparisons .............................................................................................. 23 Figure 11. SCRIM vs. CT Meter comparisons ........................................................................................... 24 Figure 12. Limits of agreement with upper and lower bounds for the Ames system and the CT Meter (Concrete 4 removed) ................................................................................................................................. 25 Figure 13. Limits od agreement with upper and lower bounds for the SCRIM system and the CT Meter (Concrete 4 removed) ................................................................................................................................. 25 Figure 14. SCRIM and Ames comparisons: Concrete 4 removed .............................................................. 26 Figure 15. Limits of agreement with upper and lower bounds for the SCRIM and Ames MPD differences .................................................................................................................................................................... 27 Figure 16. Ames and CT Meter orthogonal regression analysis: Concrete 4 removed............................... 28 Figure 17. SCRIM and CT Meter orthogonal regression analysis: Concrete 4 removed ........................... 29 Figure 18. SCRIM and Ames orthogonal regression analysis: Concrete 4 removed .................................. 30 viii CHAPTER 1: INTRODUCTION Pavement macrotexture is an important pavement surface property that affects many vehicle- pavement interactions, including friction, splash and spray generation, and tire/road noise, among other aspects. While many pavement characteristics are important to road safety,t ire pavement friction is particularly important for drivers when they need to react quickly. The frictional forces between the pavement and the vehicle’s tires help drivers maintain control of the vehicle in instances where sudden emergency maneuvers are needed . However, the level of friction generated between a tire and the road changes with the speed of the vehicle, and pavement macrotexture is the main road surface characteristic that affects this change (Fulop et al. 2000) . Both macrotexture and microtexture influence the available friction and consequently the crash risk on the roadways. Macrotexture is t he spaces between the aggregate comprising the surface of the pavement. Microtexture describes the asperities in the aggregates. This is the voids and divots found on the surface of a piece of aggregate (Rezaei et al. n. d.). In wet pavement conditions, macrotexture plays a key role in providing channels for the water to escape from under the tire . With high macrotexture, there are larger crevices in the pavement where the rain water can drain into. With low macrotexture, the holes get filled quicker and water lays on the surface of the pavement, reducing the contact between the tire and the road and consequently the friction . Therefore, if the macrotexture is low and the pavement is wet , a vehicle will need more time and a longer distance to slow down or come to a complete stop (Pulugurtha et al. 2012). Due to pavement texture ’s influence on safety, highway agencies are starting to monitor the macrotexture of the roadways at the network level. Therefore, there is a need to standardize the way macrotexture data is collected at high speeds. Historically, pavement macrotexture on roads has been measured using the sand patch test. In this test, one must take a 24.6 ml container and fill it with sand. Then, the sand is dumped on the pavement surface and a rubber disk is used to spread the sand in a circular motion. The diameter of the resulting circle is used in an equation to compute the mean texture depth (MTD) (TxDOT 2008). This process is very time consuming and operator -dependent. Current equipment used to measure the macrotexture use high -speed laser sensors. Examples include the Circular Track Meter (CT Meter) and high-speed laser-based systems, such as those found in the Sideways-Force Coefficient Routine Investigation Machine ( SCRIM) and Ames systems used in this thesis . The CT Meter is a stationary test while the SCRIM and Ames system s continuously take measurements along the length of the pavement. 1 Problem Statement The American Society of Civil Engineers (ASCE) infrastructure report card recently gave our country’s roads a D, which indicated poor conditions. ASCE reports that road safety conditions take part in about one-third of the United States traffic-related deaths. To mitigate this problem, significant resources have been invested over the years to “… add… or improve… median barrier systems, and widen… lanes and shoulders…” (ASCE 2013). While this, in conjunction with improvements that have made vehicles safer, has helped reduce the number of crashes and fatalities by 24% from 2005 to 2010, more work needs to be done. Safety can be improved by providing adequate pavement friction. With limited funds, investments must target projects with the most significant impact on the safety for drivers and their passengers. Past studies (e.g., Flintsch et al. (2003) and Pulugurtha et al. (2012)) concluded that macrotexture is an important road surface property that affects safety, making it a crucial quality to consider in pavement evaluation. Although high-frequency lasers have facilitated the collection of pavement macrotexture at highway speed, available measuring devices do not always result in identical macrotexture values. Objective The objective of this thesis is to assess the repeatability and reproducibility of two-high-speed systems used to measure macrotexture, using the static CT Meter measurements as the reference. Significance/ Contribution The CT Meter is an advancement to the sand patch test, but this method still takes time and require traffic control . Newer laser-based macrotexture measurement devices can be used in continuous stretches to collect network-level data quicker. This thesis evaluated two high-speed laser-based systems for repeatability and reproducibility. The availability of repeatable and reproducible high-speed macrotexture measurement systems is expected to facilitate, among other things, setting a standard procedure for measuring macrotexture at the network level and defining desired macrotexture levels that can be applied and maintained for roadway safety. Furthermore, if it is proven that the high-speed systems are just as reliable as the CT Meter, the newer systems can be employed. This will increase data collection speed and safety of the data collection personnel and reduce data collection costs . 2