Laboratory and field evaluation of asphalt pavement surface friction resistance

Abstract

Pavement surface friction is a significant factor for driving safety and plays a critical role in reducing wet-pavement crashes. However, the current asphalt mixture design procedure does not directly consider friction as a requirement. The objective of this study was to develop a surface friction prediction model that can be used during a wearing course mixture design. To achieve the objective, an experimental study was conducted on the frictional characteristics of typical wearing course mixtures in Louisiana. Twelve wearing course mixtures including dense-graded and open-graded mixes with different combinations of aggregate sources were evaluated in laboratory using an accelerated polishing and testing procedure considering both micro- and macro texture properties. In addition, the surface frictional properties of asphalt mixtures were measured on twenty-two selected asphalt pavement sections using different in situ devices including Dynamic Friction Tester (DFT), Circular Texture Meter (CTM), and Lock-Wheel Skid Trailer (LWST). The results have led to develop a procedure for predicting pavement end-of-life skid resistance based on the aggregate blend polish stone value, gradation parameters, and traffic, which is suited in checking whether the selected aggregates in a wearing course mix design would meet field friction requirements under a certain design traffic polishing.

Keywords

friction skid resistance / polishing / PSV / LWST / micro-texture / macro-texture

Introduction

In recent decades, the need to travel has been increased exponentially due to immense economic growth and social advancement. The number of motor vehicles operating on highways has also increased significantly, resulting in unprecedented levels of risk for highway users. According to the National Highway Traffic Safety Administration (NHTSA), during the year of 2010, 32,999 traffic crashes were fatal, 3.9 million citizens were injured, and 24 million vehicles were damaged in the United States [¹]. According to the National Transportation Safety Board (NTSB), approximately 13.5 percent of fatal crashes and 25 percent of all crashes occurred when pavements were wet [²].

Despite the complexity of highway crashes, the factors associated with those crashes can be summarized into three main categories: driver related, vehicle related, and highway condition related [³]. Out of these three categories highway agencies can control only highway conditions such as: pavement surface characteristics, pavement geometry, traffic systems, etc. One of the major reasons behind highway crashes is the low friction between the vehicle tire and the pavement surface, especially when pavement surface is wet. This has led to the interests at both the Federal and state level in reducing the occurrence of wet-pavement crashes through establishing better friction design guidelines and providing surface mixtures with adequate levels of friction (or skid) resistance over the life of the pavement [⁴].

Skid resistance is the pavement friction force that resists sliding of vehicle tire on pavement surfaces. In the United States, the most commonly used device for measuring the in-situ pavement skid resistance is called the Lock-Wheel Skid Tester (LWST). According to the American Society for Testing and Materials (ASTM) Standard E274, the LWST test measures the pavement skid resistance as the skid number (SN) at a given skid speed.

There are four major groups of factors contributing to develop friction at the tire-pavement interface: pavement surface characteristics, vehicle operating parameters, tire properties, and environmental factors. To develop better friction design guidelines for a wearing course asphalt mixture design, many studies have been attempted to establish relationships between pavement friction and surface characteristics, especially between friction resistance and the texture of pavement surface [⁵].

Pavement surface texture can be characterized by the surface asperities, measured as the deviation of the surface from true planar surface [⁶]. Based on the measured surface asperities in terms of wavelength and amplitude, the pavement surface texture can be further categorized into three levels: mega-texture, macro-texture and micro-texture [⁷]. Out of these three types of texture, the macro and micro-textures are the predominant features and both influence the change in skid resistance with vehicle speeds [⁸]. Specifically, the micro-texture is significant at low vehicle speed as it can cause adhesion between tire and the pavement surface; whereas, at high vehicle speed the macro-texture is responsible for the hysteresis friction and influences the skid resistance by increasing the friction-speed gradient and facilitating the drainage of water [⁵,⁹,¹⁰].

For flexible pavements, the micro-texture is mainly affected by the surface texture of the coarse aggregate [⁵,¹¹], and the pavement’s micro-texture is associated with the aggregate particle’s microscopic feature and refers to the small scale texture on the surface of an aggregate particle. On the other hand, the pavement macro-texture refers to the large scale texture defined by shape and size of aggregate particles presented on the road surface, which is largely varied with surface mixture types or aggregate gradations. Therefore, designing a wearing course mixture so that the pavement surface will have adequate friction over its life involves identifying an appropriate combination of micro- and macro- textures and also requires a balanced understanding between economic and engineering tradeoffs associated with selecting different mix and aggregate material types [⁵].

Objective and scope

The objective of this study was to develop a surface friction prediction model used in a wearing course mixture design that considers both polished stone value and mixture type alike in terms of micro- and macro- surface textures. To achieve the objective, an experimental study including both laboratory and field measurements on the frictional characteristics of typical wearing course mixtures in Louisiana was conducted.

Friction experimental design

Laboratory evaluation

Materials

Twelve typical asphalt wearing course mixtures including two aggregate sources (a crushed sandstone and a siliceous limestone) and four mix types were considered in the laboratory friction evaluation. Both aggregates are commonly used in the wearing course mixture construction in Louisiana, and it is known that the selected sandstone has a significantly better polishing resistance than the selected limestone.

The selected four mix types consists of a 19-mm Superpave Level-II mix, a 12.5-mm Superpave Level-II mix, a stone matrix asphalt (SMA) mix and an open graded friction course (OGFC) mix. For each mix type considered, a typical job mix formula (JMF) that is currently used in the wearing course construction projects was acquired and used to design three hot mix asphalt (HMA) mixtures using three coarse aggregate blends (i.e., aggregates retained on the 2.36mm sieve): 100 percent limestone, 100 percent sandstone and a combination of 70 percent limestone and 30 percent sandstone. More details of the mix design information may be referred to elsewhere [¹¹].

Laboratory tests

Polish Stone Value (British Pendulum) tests were chosen to evaluate the polishing resistance for the selected aggregates in this study. Since current HMA specifications do not provide any standard friction test procedure during mix design, a polishing and friction testing procedure developed at the National Center for Asphalt Technology (NCAT) for rapidly evaluating the frictional performance of HMA mixtures [¹²] was selected in this research study. Brief description for each test considered and testing procedures are provided below.

Polish Stone Value (PSV) test

This test follows both testing procedures of AASHTO T 278 and T 279 and is a two stage test. First, aggregate coupon samples are made by fixing coarse aggregates into slightly curved coupon molds through applying an epoxy binder to flat surface of selected aggregates. Then, those coupon samples are installed around a testing wheel and subjected to an accelerated polishing action in a special polishing machine for 10 hours (Figure 1). The state of polish reached by each coupon sample is then tested with a British Pendulum Tester (BPT) by swinging the pendulum with a specific normal load and standard rubber pad over the aggregate surface. The average number of the BPT results (so-called British Pendulum Number, or BPN) is reported as the polish stone value (PSV) for the tested aggregate. The higher the PSV, the more polish resistance of the aggregate.

Accelerated polishing device

As shown in Figure 2, the accelerated polishing machine developed at the NCAT is called the Three Wheel Polishing Device (TWPD) designed to simulate the traffic-polishing effects on surface friction characteristics of asphalt mixtures by using a three abrasion wheels assembly. This machine is capable of polishing a donut-shape area with a mean diameter of 280 mm (11.2 in) on the surface of a lab-made testing slab with a dimension of 50 cm (20 in) by 50 cm (20 in). The normal load during the test is 468 N (105 lb.) with tire pressure of pneumatic tires maintained at 2.9 MPa (50 psi). During the slab polishing, water is continuously sprayed to simulate a wet polishing in the field. It was found that such polishing device together with a set of friction/texture measurements could be used to evaluate the frictional resistance of HMA mixtures in the laboratory that represents field measured results [¹²].

Polishing and friction measurement procedure

As mentioned above, the TWPD test requires the preparation of testing slabs with a planar dimension of 50 cm (20 in) by 50 cm (20 in). The slab thickness is usually 75 mm (3 in). In this study, three replicate slabs were prepared for each of the twelve HMA mixtures considered. A kneading-compaction procedure was followed with a target air void of 7 percent. Details of the preparation of friction testing slabs as well as compaction procedures can be referred to elsewhere [¹¹].

Each slab was polished under the TWPD device for a maximum of 100,000 polishing cycles. The slab surface friction and texture properties were measured at predefined polishing cycles. Specifically in this study the predefined polishing cycles are 0 (before polishing), 5, 10, 30, 50, and 100 thousand cycles. At the end of each predefined polishing interval the accelerated polishing device was stopped and the slab was removed and dried for the evaluation of its surface texture by a circular track meter (CTM) and its friction resistance using a dynamic friction tester (DFT).

Dynamic friction test (DFT)

As described in ASTM E1911, a DFT device includes a motor and a horizontal spinning disk that is attached with three rubber sliders spring-mounted at a diameter of 280 mm (11.2 in). The disk is initially suspended above the measurement surface and lowered down only when a desired tangential speed of the sliders is attained. As the three rubber sliders start to contact the measurement surface, the disk is spinning down. During the spin down, by measuring the friction force in each slider with the corresponding speed the coefficient of friction of the surface is determined. Note that water is sprayed to the measurement surface during the testing. The DFT system can be used to measure the friction at a speed over the range of 0 to 90 km/h (0 to 55 mph) [¹²]. In this study, the DFT measured friction numbers at 20, 40 and 60 km/h were used in the analysis.

Circular texture meter (CTM)

As described in ASTM E 2157, the CTM equipment consists of a laser displacement sensor that is mounted on a motor-driven rotation arm. The profile measured under the laser displacement sensor follows a circle of 280 mm (11.2 in) diameter. The result is the mean profile depth (MPD) for the surface under consideration, which is representative of surface macro-texture.

Field evaluation

Twenty two field test sections were selected across Louisiana to cover the typical wearing courses; Superpave [12.5 mm and 19mm], SMA, and OGFC. A 305-m (1000-ft) long straight section on each project was marked without sharp curve, steep grade and intersection for friction test. In each section, LWST were run at 72 km/h (40 mph) in two passes, one with the smooth tire locked and the other with the ribbed tire locked. Three skid number measurements were taken at the beginning (0 m), the mid-point (152.5 m), and the end (305 m) of each test section. Three DFT and three CTM tests were conducted within each segment at 4ft. interval where LWST takes the skid number reading. The DFT and the CTM were run at exactly the same spot. The layout and the locations of test spots are shown in Figure 3.

Lock-wheel skid trailer (LWST)

LWST is the most common field friction test device in the United States that is able to test skid resistance at normal traveling speeds. The ASTM standard for friction devices using full-scale test tire was followed during the test which is ASTM E 274. ASTM E 501 for a ribbed tire and ASTM E 524 for a smooth tire were followed. Since the test tire is fully locked during the test, the slip speed of the tire on the pavement is equal to the traveling speed of the test vehicle. Most of the time, LWST was operated at a speed of 72 km/h (unless specified, although other speeds may be also used). The output of the LWST test is the measured skid number (measured coefficient of friction multiplied by 100). The skid number was reported in the form of SN|test speed|tire type.

Laboratory test results

PSV test results

The British Pendulum and aggregate accelerated polishing tests were used to measure the PSV of coarse aggregates considered in both laboratory study and the twenty-two selected pavement projects. PSV of individual aggregates are presented in Table 1. Note that the Louisiana Department of Transportation and Development (LADOTD) maintains an “Approved Materials List” that conforms to the quality requirements of the Department’s Standard Specifications for Roads and Bridges for each construction material produced or supplied. The approved list of pavement aggregates, which is used to provide quality material sources in a mixture design, only contains “Product Source Code” (i.e., producer/supplier sources) and approved aggregate friction ratings based on the PSV measurement. Since lots of projects selected use blending of different aggregates in mixtures, a final PSV of those surfaces was determined by weighted average of individual aggregate in the mixture. A term blend PSV was defined as the single PSV of mixture containing more than two coarse aggregates based on the proportion percentages of individual coarse aggregates contained in the mix.

DFT test results

During the laboratory testing, the DFT measured surface friction results are expressed in terms of percentage values called as the dynamic friction numbers (DFN), which are the DFT measured coefficients of friction multiplied by 100. In general, the following observations were made from the DFT test results.

Figure 4 shows that the mean DFN values increased with the decrease of sliding speeds and decreased with the increase of polishing cycles. In addition, test variations under three different sliding speeds (the error bars in the figure) showed a similar trend. Figure 4 also displays that the peak DFN friction values usually occurred between 5,000 and 10,000 polishing cycles rather than at the beginning of the testing, primarily due to the development of an early surface roughness or textures of the coated aggregate particles (e.g. remove the excess binder from the surface and expose the aggregate) [⁴].

Figure 5 presents the mean DFN results measured at 20 km/h (DFN₂₀) on polished slabs of the three OGFC mixtures considered. The DFN₂₀ curves clearly indicate that the sandstone OGFC mixture performed the best in terms of friction resistance among the three OGFC mixtures considered. Figure 5 also shows that the OGFC mixture with a combination aggregate blend (30 percent sandstone and 70 percent limestone) also had better friction resistance than the mixture with 100 percent limestone. Similar DFT results were obtained for other mix types considered.

Figure 6 presents a comparison of DFT₂₀ results for the four sandstone asphalt mixtures considered. As can be seen in the figure, no single mixture could show a distinct DFN₂₀ curve as compared to the others. As different mix types indicate different macro-textures, such “no distinct difference” friction measurement results simply implied that the DFT is better used as an indicative test for the surface micro-texture, not macro-texture of mixtures.

CTM test results

Figure 7 presents the CTM results in this study. The CTM results were reported as the mean profile depth (MPD) in the unit of mm. A high MPD value represents high surface macro-texture. Noted that each solid line in Figure 7 shows the variation of the mean MPD results for each mix type with the number of polishing cycles; whereas, the associated error bars represent the MPD variations due to the change of different aggregate blends (i.e., 100 percent sandstone, 100 percent limestone and the combination aggregate blend considered). In general, except two Superpave mix types, the CTM results clearly showed distinctly different MPD values for different mix types evaluated. That is, the OGFC mix showed the highest MPD values (as having the highest surface macro-texture), followed by the SMA mix and the two Superpave mixes. The MPD values under different polishing cycles tended to remain as a constant after the initial 2000 polish cycles. This indicates that the MPD values are un-affected by the polishing. The initial variation in the MPD measurement could be related to aggregate abrasion during polishing or experiment errors [⁴]. The overall CTM test results confirmed that the CTM measured MPDs are more indicative to the surface macro-texture (due the change of the mix type and gradation) than the surface micro-texture (related to the aggregate type).

Prediction of F(60)

As discussed in the literature, the friction or skid resistance offered by an asphalt surface is related to both surface micro- and macro-textures. To better quantify a joint influence by both micro- and macro- textures, a standardized friction number called the International Friction Index (IFI) was introduced by the Permanent International Association of Road Congress (PIARC) [¹³]. The IFI consists of two numbers: speed constant (S_P) and friction number F(60). While the S_p is a parameter directly related to macro-texture of the road surface and normally needed in shifting a friction value to the standard test speed, the F(60) represents a standardized wet surface friction number at 60 km/h. In this study, the F(60) values of wearing course mixtures were computed using the following Equation 1 as specified in ASTM E 1960. Note that a pair of the MPD and DFN₂₀ in Eq. (1) may be viewed as an indicative measure of macro- and micro- texture of an asphalt surface considered.

(1)\mathrm{F}(\mathrm{60})=\mathrm{0.081}+\mathrm{0.732}{\mathrm{DFN}}_{\mathrm{20}}\times e^{\frac{-\mathrm{40}}{\mathrm{14.2}+\mathrm{89.7}\mathrm{MPD}}}

The computed F(60) values for the twelve HMA mixtures in the laboratory study are shown in Figure 8. As expected, the F(60) values generally decrease with the increase of polishing cycles. The sandstone OGFC and SMA mixtures showed having the highest F(60) values (due to both having the greatest micro- and macro- textures), followed by (in a descending order) the sandstone Superpave mixtures, limestone SMA, and the combination aggregate blends of OGFC and SMA. Lastly, because of the combination of low micro- and macro- textures the two Superpave mixtures of 100 percent limestone were showed to have the lowest F(60) values.

Field testing results

The selected field projects contain eight coarse aggregate sources and four mixture types commonly used in Louisiana were covered in this study. The project information (e.g. project number, pavement age, aggregate percentage, and traffic ADT) together with the average field friction measurement results are summarized in Table 2. The pavement ages are ranged from newly constructed to sixteen years old. Traffic-wise, the ADT numbers of selected projects are varied from 400 to 36,000 vehicles (i.e., low-volume roads to high-volume roads).

Skid prediction procedure

After the collection of laboratory and field friction data, several correlation studies were performed. Based on the correlation analysis a skid prediction procedure was developed. Figure 9 presents the skid prediction procedure proposed in this study by considering aggregate property (PSV), mixture property (i.e., gradation), and traffic and discussed below.

In order to address the effect of traffic polishing on pavement surface friction, a term Traffic Index (T.I.) was defined. This is a measure of total number of traffic plied on the pavement surface by considering ADT and traffic growth rate factor as given in Eq. (2).

(2)\mathrm{Traffic\; Index}(\mathrm{T}.\mathrm{I}.)={\displaystyle \frac{\mathrm{ADT@design}\hspace{0.17em}\mathrm{lane\times Growth}\hspace{0.17em}\mathrm{rate} \hspace{0.17em}\mathrm{fatcor\times 365}}{{\mathrm{10}}^6}}

PSV is a measure of aggregates frictional property (micro-texture). A DFT measurement value at speed 20km/hr. is also considered as the measure of pavement surface micro-texture. From the analysis of variance it was found that the primary sources of variation of DFT measurements were aggregates and traffic, and both correlations between DFT20 vs. traffic (in terms of T.I) and between DFT vs. PSV followed an exponential trend with high R-value of 0.95 and 0.91, respectively. After several trial regression analyses in SAS, from aggregate’s PSV and traffic a best DFT20 prediction model is proposed as in Eq. (3).

(3)\mathrm{DFT2}\mathrm{0}=\mathrm{A}\times {\exp }^{\mathrm{(B\times T}\mathrm{.I}\mathrm{.)}}+\mathrm{C\times PSV+D\times }{\exp }^{\mathrm{(PSV)}}(R^{\mathrm{2}}=\mathrm{0.88})

where A= 0.13, B= –0.056, C= 2.6 and, D= –0.5 are regression coefficients.

In the above equation PSV is divided by 100.

Pavement surface macro-texture is strongly related with asphalt mixture properties. The CTM measured Mean Profile Depth (MPD) are commonly referred as a measure of surface macro-texture. The MPD prediction model proposed by Masad et al. [¹⁴] was adopted in this study. As shown in Figure 9 the MPD were estimated based on mixture gradation. First the parameters K and \lambda were determined by fitting the gradation curve in Weibull distributions (Eq. (4)). Then a prediction equation (Eq. (5)) to predict MPD was used.

(4)F(x:K,\lambda )=\mathrm{1}-{\exp }^{{(-\frac{x}{\lambda })}^K}

(5)\mathrm{MPD}=\mathrm{0.14}\times \lambda +\mathrm{0.09}\times K-{\displaystyle \frac{\mathrm{0.041}}{K^{\mathrm{4}}}}

where x= Aggregate size in milimiters, K = Shape factor of Weibul distribution, \lambda = Scale factor of Weibul distribution, MPD= Mean profile depth measured by CTM.

It was found that smooth tire skid number can closely correlate the actual field friction than ribbed tire. Since DFT and CTM are the measure of micro and macro textures, several trial models were performed in SAS and a best fit nonlinear regression correlation among SN40S, DFT20 and MPD is proposed as given by Eq. (6).

(6)\mathrm{SN40S}=\mathrm{2.15}\times \mathrm{DFT}\mathrm{20}\times {\exp }^{(-\frac{\mathrm{0.54}}{\mathrm{MPD}})}(R^{\mathrm{2}}=\mathrm{0.73})

where SN40S= Skid number at 40 mph using smooth tire divided by 100. MPD= Mean profile depth measured using CTM. DFT20= DFT reading at speed 20 km/h.

Similarly, for given SN40S and MPD a SN40R prediction model was also developed as given in Eq. (7) with the R² value of 0.75.

(7)\mathrm{SN40R}=\mathrm{0.93}\times \mathrm{SN40S-0}\mathrm{.16}\times \mathrm{MPD+0}\mathrm{.26}

All the procedures presented from Eqs. (2) to (7) should be used to predict the field skid number from aggregate PSV, gradation, and traffic. The system provided in the flow chart (Figure 9) can also be used to evaluate the aggregates in the mixture and determine whether they are suitable to achieve desired skid number. Starting from the design skid number for given mixture type and traffic level and performing the back calculations, optimum aggregate’s PSV can be determined. To select the optimum combination of aggregate type for given mixture in order to achieve the desired level of skid resistance during a wearing course mix design, following steps should be followed:

• Determine the friction demand for a specific mix design and select a design skid number at the end of design life (e.g., SN40S= 20).

• Compute the design traffic index using equation.

• Select a mixture type (i.e., 19 mm or 12.5 mm Superpave, SMA, and OGFC) with aggregate gradation.

• Calculate \lambda and K from the selected aggregate gradation using equation.

• Predict the macro-texture (MPD) for the mixture considered using calculated \lambda and K.

• Back-calculate the required DFT20 at the end of design life (the minimum allowed DFT20 value) using design skid number and predicted MPD.

• Predict a required micro-texture, or PSV_req using calculated DFT20 and traffic index.

• Choose a coarse aggregate blend used in the mix that has a blend PSV value higher than PSV_req.

To demonstrate an actual accuracy of the proposed skid prediction model aforementioned, thirteen field pavement projects were further identified from the Department’s Pavement Management System (PMS). The detail information for each project selected can be referred elsewhere [⁹]. First, the DFT and MPD for each project were calculated using PSV, traffic, \lambda, and K data. Then using calculated DFT20 and MPD values, the SN40S of each project were calculated using Eq. (6). Figure 10 presents a comparison plot between calculated and field measured skid numbers. As seen in Figure 10, the predicted SN40S values generally matched fairly well with measured ones for the pavement projects selected.

Summary and conclusions

A suite of accelerated slab-polishing and friction tests on twelve wearing course mixtures containing four mix types and three aggregate blends were conducted in this study. In addition, twenty-two asphalt pavement test sections covering a wide range of material type and traffic conditions were selected to measure surface texture and skid resistance properties. The collected data and measurements including series of laboratory and field friction measurements, such as PSV, DFT, CTM and LWST, were used to perform comprehensive statistical analyses of the influence of aggregate, mixture design and traffic on pavement surface friction resistance. The following observations and conclusions can be drawn from this study:

• The DFT test results indicated that all sandstone mixtures (irrelevant of mix type) had significantly higher friction resistance than their counterpart mixtures with limestone. In addition, mixtures with the combination aggregate blends were also found to have better friction resistance than those limestone-only mixtures.

• The CTM test results revealed that the OGFC mix type generally had the highest surface macro-texture among the four mix types investigated, followed by the SMA, and then by the two Superpave mix types. The two Superpave mix types in this study were found to have similar levels of macro-texture, primarily due to having similar dense-graded aggregate gradations.

• Analysis of mixtures’ skid resistance as quantified by the International Friction Index friction number F(60) indicated that it is generally possible to design an asphalt mixture with overall satisfactory skid resistance through blending of low and high skid-resistant aggregates (as specified by the PSV test) into a mix design.

• A new pavement end-of-life skid prediction methodology was proposed considering PSV, gradation parameter (K and l), and traffic. A correlation between DFT20 and PSV was established including the traffic, the MPD values were predicted from gradation parameters K and l, and a correlation was established among SN40S, DFT20 and MPD. A linear regression analysis was performed to predict SN40R from SN40S and MPD. Finally, a procedure to predict the field skid number from aggregate and mixture properties along with traffic was established. The skid prediction procedure can be used by engineers in the asphalt design stage to achieve desired end-of-life skid number at field.

• The proposed skid prediction procedure generally shows a great potential in analyzing the PMS skid number data and developing a set of threshold (or minimum) skid number values for roads under different traffic and design speeds. Future research may be also towards defining a unified, statewide skid number testing results that can provide the guidance in a routine wearing course mix design.

References

[1] Blincoe L J, Miller T R, Zaloshnja E, Lawrence B A. The Economic and Societal Impact of Motor Vehicle Crashes, 2010. Report No. DOT HS 812 013.Washington, DC: National Highway Traffic Safety Administration, 2014

[2] Kuemmel D A, Sontag R C, Crovetti J A, Becker Y, Jaeckel J R, Satanovsky A. Noise and Texture on PCC Pavements. Final report multi state study. Report Number WI/SPR-08-99, Wisconsin Department of Transportation, Kinsman Blvd., Madison, WI, 2000

[3] Noyce D A, Bahia H U, Yambό J M, Kim G. Incorporating Road Safety into Pavement Management: Maximizing Asphalt Pavement Surface Friction for Road Safety Improvements. Draft Literature Review and State Surveys, Midwest regional University Transportation Center, Traffic Operations and Safety (TOPS) Laboratory. Apr. 29, 2005

[4] Masad E, Rezaei A, Chowdhury A, Harris P. Predicting Asphalt Mixture Skid Resistance Based on Aggregate Characteristics. Texas Transportation Institute, The Texas A&M University System College Station, Texas, 77843, 2009

[5] Hall J W, Smith K L, Titus-Glover L, Wambold J C, Yager T J, Rado Z. Guide for Pavement Friction. NCHRP Project 1-43. National Cooperative Highway Research Program, Transportation Research Board, National Research Council, Washington, D.C., 2009

[6] Dewey G R, Robords A C, Armour B T, Muethel R. Aggregate Wear and Pavement Friction. Presented at 80th Transportation Research Board Annual Meeting, Washington, D.C., 2001

[7] Hogervorst D. Some properties of crushed stone for road surfaces. Bulletin of Engineering Geology and the Environment, 1974, 10(1): 59–64 (Springer)

[8] Rose J G, Gallaway B M. Macrotexture Measurement and Related Skid Resistance at Speeds from 20 to 60 Miles per Hour. Transportation Research Record 341, Transportation Research Board, TRB, National Research Council, Washington, D.C., 1970

[9] Ashby J T. Blended Aggregate Study. LTRC final report 145, Baton Rouge, 1980

[10] Bloem D L. Skid-Resistance: The Role of Aggregates and Other Factors. National Sand and Gravel Association Circular 109, Silver Spring, MD, 1971

[11] Wu Z, King B. Developments of Surface Friction Guidelines for DOTD FHWA/La.011/485. Louisiana Transportation Research Center, Baton Rouge, 2012

[12] Vollor T W, Hanson D I. Development of Laboratory Procedure for Measuring Friction of HMA Mixtures-Phase 1. NCAT Report 06-06, National Center of Asphalt Technology, Auburn University, AL, 2006

[13] Wambold J C, Antle C E, Henry J J, Rado Z. PIARC (Permanent International Association of Road Congress) Report. International PIARC Experiment to Compare and Harmonize Texture and Skid Resistance Measurement, C-1 PIARC Technical Committee on Surface Characteristics, France, 1995

[14] Masad E, Rezaei A, Chowdhury A, Harris P. Field Evaluation of Asphalt Mixture Skid Resistance and Its Relationship to Aggregate Characteristics. Texas Transportation Institute, The Texas A&M University System College Station, Texas 77843, 2010

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature