Search
Close
Search
Advanced Search
Search Menu
AI Discovery Assistant

Abstract

Pavement skid resistance is critical for contributing to traffic safety. In this study, two pavement specimens were fabricated using different aggregate gradations and binder contents. Compaction efforts were made to ensure the same porosity of the two specimens. Some parameters related to the surface morphology were measured using the sand patching method and a 3D laser. The parameters include mean texture depth (MTD), interface roughness (Ra), skewness (Rsk), kurtosis (Rku), fractal dimension (D) and parameters related to material ratio curve. Relationships between these parameters and skid resistance were quantitatively developed. Skid resistance increased along with the increase of MTD, Raand D. From the surface A to surface B, MTDincreased from 0.23 to 1.10 mm, Ra increased from 0.93 to 1.80 mm, Dincreased from 2.25 to 2.43 and the British Pendulum Number increased from 80 to 97 accordingly. The comprehensive analysis could assist scholars and practitioners to facilitate the understanding of tyre-pavement interaction and further assist pavement design.

Issue Section:
Research article

1. Introduction

Skid resistance of a pavement is the force that resists a tyre from rotating on the surface of a pavement. Skid resistance is a critical aspect in pavement design due to the fact that adequate skid resistance would reduce the occurrence rate of skid-related traffic accidents. Skid resistance of a pavement is affected by many factors, including pavement surface characteristics, tyre performance, vehicle parameters and environmental factors. It has been reported that an obvious slipperiness would result in a high crash rate, while a slight slipperiness would lead to a low crash rate [1]. It was reported that there exists a dramatic increase in the occurrence rate if the friction performance decreased under certain threshold values [2]. Icy and snowy weather conditions could noticeably increase the casualty rate by 25% [3]. When the friction index (SCRIM) increased from 50 to 60, the reduction of wet-pavement accidents can be up to 68% [4]. After the conduction of maintenance, skid resistance is also a critical concern to ensure driving safety [5].

Due to the fact that the skid is caused by the pavement-tyre interaction, the surface characteristics of pavement are key factors impacting the skid resistance. Microtexture and macrotexture of a surface remarkably affect the skid resistance [6,7]. Microtexture is the fine-scale texture with wavelengths from 0 to 0.5 mm, closely related to the surface morphology of aggregate. Microtexture is believed to interact with tyres on a molecular scale and provide adhesion between pavement and tyre. It was reported that microtexture played critical roles in affecting the skid resistance in low-speed and dry conditions [8]. Macrotexture refers to the surface behaviours of a pavement due to the combined effects of aggregate shape, gradation, binder content and compaction, etc. Macrotexture is generally in a scale from 0.5 to 50 mm. It impacts the skid resistance of a pavement in high-speed and wet states [8]. Moreover, driving speed plays an obvious role in skid resistance, which is manifested as a high speed generally generates a greater friction than a low speed [9]. In predicting the skid resistance of asphalt pavement, a model including both microtexture and macrotexture should be considered to give an improved prediction accuracy [10].

Many scholars have investigated the effects of other parameters relating to the texture on the skid resistance. Cong and Wang [11] found that fine aggregate angularity (FAA) had a significant impact on skid resistance on the macrotexture level but only a marginal influence was observed on the microtexture level. Li et al. [12] demonstrated that entropy, which is a textural parameter, and peak curvature imposed the most significant impact on skid resistance, while mean profile depth (MPD) demonstrated the least impact. Hu et al. [13] found there was no positive relation between MPD and dynamic friction coefficient at different speeds. The peak density (Spd) and the arithmetic mean of peak curvature (Spc) presented positive effect on the friction. Moreover, the influence introduced by Spd and Spc depended on the test speed. Din and Mir [14] found that the addition of copper slag (CS) in asphalt pavement increased the skid resistance due to the high angularity, density and friction angle of CS particles. Alsheyab and Khasawneh [15] found that the air void volume and effective binder volume exerted an obvious impact on friction resistance, which can be attributed to the fact that the air void volume could change the macrotexture and the effective binder volume could affect the microtexture.

The skid resistance attenuates during the service under the coupling effect of traffic loading and environmental factors. Plati and Pomoni [16] stated that pavement geometric characteristics and heavy traffic are the most critical factors affecting the skid resistance and macrotexture evolution rather than the total traffic flow. Exceptions existed that during the service of asphalt pavement, MPD increased and skid resistance decreased. This is mainly ascribed to the fact that the fine aggregate in the surface was first polished by traffic before the polish of the aggregate, which changed the microtexture, induced a larger MPD and a reduced skid resistance. The measurement of aforementioned macro and micro parameters sometimes requires a lot of manpower and material resources. In this scenario, some scholars proposed new parameters or new predicting models for evaluating the skid resistance. Chou et al. [17] proposed a parameter-mean difference of elevation (MDE), the measurement of which could avoid the time-consuming and other difficulty when measuring traditional indicators, such as mean texture depth (MTD) and MPD, etc. Kane and Edmondson [18] defined a novel parameter termed ‘averaged aggregate hardness parameter (AAHP)’ to characterize the skid resistance of aggregate. Results showed that AAHP correlated well with the long-term friction performance of asphalt pavement. Based on fluid dynamics theory and solid mechanics theory, Fwa and Chu [19] proposed a concept of skid resistance state to explore the interaction behaviours between tyre and pavement with the incorporation of water film on pavement surface. Rezaei et al. [20] built a predicting model as a function of traffic level, microtexture, and aggregate gradation. This model proved to be accurate in predicting the skid resistance of asphalt pavements under various conditions.

It is generally believed water reduces the friction resistance [21]. Dan et al. [22] measured the skid resistance of asphalt pavement under the effect of slipperiness. Results indicated that the ice pavement under wet and low-temperature conditions led to the worst pavement skid resistance. Moreover, with increase in ice thickness, the texture depth would be gradually covered by ice. In this process, the skid resistance would gradually decline. After the texture depth was fully covered, the thickness increase of ice may not have a significant effect on the skid resistance. Tang et al. [23] built a tyre-water-pavement interaction finite element analysis (FEA) model. This model could simulate the skid resistance of wet porous asphalt pavement. It revealed that a pavement with a larger MPD values and a higher porosity and a tyre with an appropriate tread pattern could increase the skid resistance under wet conditions. The higher porosity could allow quick and effective water drainage from the pavement surface, and the larger MPD and the appropriate tread pattern significantly strengthen the pavement-tyre interaction. Due to this phenomenon, porous pavements, including open-grade friction course (OGFC) [24,25,26] and permeable concrete pavement (PCP) [27], are widely used nowadays to enhance driving safety on rainy days. These pavements are featured by the large content of interconnected air voids, which allow effective water drainage. Some field tests showed MPD values of OGFC were significantly larger than those of dense-graded asphalt [28]. The friction number (FN) of OGFC was generally larger than FN in stone mastic asphalt (SMA) pavements or the superpave dense-graded asphalt pavement [29].

The aforementioned studies gave lots of valuable findings on the relationship between various surface parameters and the skid resistance of asphalt pavement. However, besides of the general parameters, such as MTD, MPD, etc., there are other parameters to characterize the surface topography, such as fractal dimension, areal material ratio, etc. Currently, the effects of these parameters on the skid resistance of asphalt pavements are not clear. Investigating the relationships between these parameters and the skid resistance of asphalt pavement would facilitate the fundamental understanding of the pavement-tyre interaction, and further could assist the pavement design. This is the motivation of this study.

2. Methodology

2.1 Materials and sample preparation

Two types of asphalt concretes (A and B) were prepared with different aggregate gradations and binder contents. The non-modified #70 binder and styrene-butadiene-styrene (SBS) modified binder were used in A and B. The asphalt contents in A and B are 6.2% and 5.84%, respectively. Table 1 and Table 2 present some basic properties of binders used in specimens A and B. Fig. 1 shows the gradations of specimen A and B, respectively. The nominal maximum aggregate size (NMAS) was 9.5 and 13.0 mm for A and B, respectively.

Aggregate gradations.
Fig. 1.

Aggregate gradations.

Open in new tabDownload slide
Table 1.
Open in new tab

Properties of #70 binder used in specimen A

ItemUnitRequirementTest resultStandard
Penetration (25 °C), 100 g·5 s0.1 mm60–8069T0604
Ductility (5 cm/min at 10 °C)cm≥15>60T0605
Ductility (5 cm/min at 15 °C)cm≥100>100T0605
Soft point°C≥4647.5T0606
Flash point°C≥260>300T0611
Solubility (TCE)%≥99.599.83T0607
ItemUnitRequirementTest resultStandard
Penetration (25 °C), 100 g·5 s0.1 mm60–8069T0604
Ductility (5 cm/min at 10 °C)cm≥15>60T0605
Ductility (5 cm/min at 15 °C)cm≥100>100T0605
Soft point°C≥4647.5T0606
Flash point°C≥260>300T0611
Solubility (TCE)%≥99.599.83T0607
Table 1.
Open in new tab

Properties of #70 binder used in specimen A

ItemUnitRequirementTest resultStandard
Penetration (25 °C), 100 g·5 s0.1 mm60–8069T0604
Ductility (5 cm/min at 10 °C)cm≥15>60T0605
Ductility (5 cm/min at 15 °C)cm≥100>100T0605
Soft point°C≥4647.5T0606
Flash point°C≥260>300T0611
Solubility (TCE)%≥99.599.83T0607
ItemUnitRequirementTest resultStandard
Penetration (25 °C), 100 g·5 s0.1 mm60–8069T0604
Ductility (5 cm/min at 10 °C)cm≥15>60T0605
Ductility (5 cm/min at 15 °C)cm≥100>100T0605
Soft point°C≥4647.5T0606
Flash point°C≥260>300T0611
Solubility (TCE)%≥99.599.83T0607
Table 2.
Open in new tab

Properties of SBS-modified binder used in specimen B

ItemUnitRequirementTest resultStandard
Penetration (25 °C), 100 g·5 s0.1 mm30–6055.6T0604
Ductility (5 cm/min at 5 °C)cm≥2030.2T0605
Soft point°C≥6078.9T0606
Flash point°C≥230>300T0611
Solubility (TCE)%≥9999.65T0607
ItemUnitRequirementTest resultStandard
Penetration (25 °C), 100 g·5 s0.1 mm30–6055.6T0604
Ductility (5 cm/min at 5 °C)cm≥2030.2T0605
Soft point°C≥6078.9T0606
Flash point°C≥230>300T0611
Solubility (TCE)%≥9999.65T0607
Table 2.
Open in new tab

Properties of SBS-modified binder used in specimen B

ItemUnitRequirementTest resultStandard
Penetration (25 °C), 100 g·5 s0.1 mm30–6055.6T0604
Ductility (5 cm/min at 5 °C)cm≥2030.2T0605
Soft point°C≥6078.9T0606
Flash point°C≥230>300T0611
Solubility (TCE)%≥9999.65T0607
ItemUnitRequirementTest resultStandard
Penetration (25 °C), 100 g·5 s0.1 mm30–6055.6T0604
Ductility (5 cm/min at 5 °C)cm≥2030.2T0605
Soft point°C≥6078.9T0606
Flash point°C≥230>300T0611
Solubility (TCE)%≥9999.65T0607

Asphalt concretes were compacted using a roller compactor. The thickness of the slabs was about 7.5 cm. Some trail tests were carried out to ensure the porosity was approximately 4%. The width and length of the samples were both 30 cm. Fig. 2 presents specimens A and B.

Specimens: (a) specimen A; (b) specimen B.
Fig. 2.

Specimens: (a) specimen A; (b) specimen B.

Open in new tabDownload slide

2.2 Surface parameters

Surface parameters used to characterize the morphology include mean texture depth (MTD), interface roughness (Ra), skewness (Rsk), kurtosis (Rku) and fractal dimension (D). Besides, some parameters related to the material ratio curve were also obtained to represent the surface behaviours.

2.2.1 Texture depth

Texture depth is an indicator of the macrotexture and presents the overall roughness of a pavement. An adequate texture depth assists braking and the water dispersion on the area between the pavement and tyres. The macrotexture of a pavement contributes greatly to the skid resistance, predominately at medium to high speeds. Meanwhile, the texture depth of the underlying layer is a significant factor contributing to the layer bonding in various asphalt pavements [30,31]. Sand patching method is commonly used to measure the texture depth.

2.2.2 Interface roughness, skewness and kurtosis

In the calculation of interface roughness (Ra), kurtosis (Rku), skewness (Rsk) and fractal dimension (D), the first step was to obtain the three-dimensional contour of the surface. A 3D laser scanner (Fig. 3) was employed to get the vertical surface morphology data, or the three-dimensional coordinates of the discrete points on a surface. The vertical coordinates of the points on the surface were obtained every 0.5 mm in length and width. After obtaining the 3D coordinates, the interface could be reconstructed using the Matlab platform.

Surface scanning.
Fig. 3.

Surface scanning.

Open in new tabDownload slide
Interface roughness (Ra), also called arithmetical mean height of a pavement surface, refers to the mean value of the discrete heights (asperities) and depths from the arithmetic mean elevation of the profile. The one-dimensional Ra could be calculated using Eq. (1). Fig. 4 shows a surface profile. If one-dimensional roughness extends to two-dimensional roughness, Eq. (2) can be used to obtain Ra.
(1)
(2)
(3)
where Sq is the root mean square of the roughness value, z(x, y) is the vertical distance between the discrete points to the mean line, as shown in Fig. 4. A is the surface area. m and n are the numbers of grids when calculating the roughness using discrete points of the surface.
Surface profile.
Fig. 4.

Surface profile.

Open in new tabDownload slide
Skewness (Rsk) is an indicator of the average of the first derivative of a surface. It represents how far the surface deviates from the plane of symmetry. A negative Rsk indicates that the surface is mainly made up of valleys, and a positive Rsk indicates the surface contains mainly peaks and asperities. Rsk equals to 0 indicates valleys and peaks are symmetrically distributed against the mean surface. Therefore, a surface with a negative Rsk is easy to be lubricated from another surface. Kurtosis (Rku) is a measure of sharpness of the roughness profile. Rku larger than 3 indicates the vertical distribution of the profile is spiked; Rku lower than 3 indicates the height distribution is skewed above the mean plane; and Rku equals to 3 indicates the height distribution is a normal distribution, which is shown in Fig. 5.
(4)
(5)

2.2.3 Material ratio curve

The profile distribution, or height distribution, can be plotted as a histogram of the vertical heights that quantify the number of the discrete points on a surface that lies at a given height range. The areal material ratio curve is the cumulative distribution of the height distribution. The ratio curve represents heights at which the areal material ratio varies in the range of 0–100%. In Fig. 6, Smr(c) is the ratio between the cross-section area of the surface at a height(c) and the reference cross-section area. This ratio is expressed as a percentage. Smr(c) can be used to determine the cutting area remaining after a certain depth of surface is removed.

By connecting two points on the areal material ratio curve with a difference of 40%, if the sum of squared deviation in the vertical-axis direction obtains the minimum value, this line is the equivalent line (Fig. 6). Some parameters relating to the material ratio curve could be used to characterize the surface morphology. In Fig. 6, after drawing the equivalent curve, the intersections between the equivalent curve and the vertical lines with 0 and 100% ratio can be determined. Then, the parameters (A1 and A2) are defined as areas of right-angled triangles. A1 and A2 can be regarded as the peak cross-sectional area and the valley cross-sectional area. In the vertical axis, Spk is the reduced peak height, representing the mean height of peaks above the core surface. Svk is the reduced valley depth, representing the mean depth of valleys underneath the core surface.

In addition, some studies also employed other volume-related parameters to characterize the areal material ratio curve. Similar to Fig. 6, a surface composes of peak, core and valley (Fig. 7). Areal material ratio 10% is regarded as the break point between the peak and core parts, while material ratio 80% is the point separating the core and dale regions. In Fig. 7, Vmp represents the volume at areal material ratio 10%. Vmc represents the material volume difference between the two material ratios 10% and 80%. Vvc represents the void volume difference between the two material ratios 10% and 80%. Vvv represents the void volume of dale at areal material ratio 80%.

2.2.4 Fractal dimension

Pavement surface presents self-similarity, indicating the pavement texture can be characterized by fractal theory [32]. Fractal dimension is generally used to approach the description of the true attributes and states of complex systems. Xie et al. [33] proposed a projective covering method for estimating the fractal dimension (FD) accurately. When using the method to obtain FD, the projective surface was divided with grids with different sizes. Although some other methods, such as the box-counting method [34], differential box counting method [35], etc., the projective covering method is more straightforward. Therefore, the projective covering method is used in this study. In Fig. 8, when the kth square abcd (a, b, c and d are the four points of a square) with a selected scale of δ×δ were selected in the reference plane, the heights, or the lengths (Aa, Bb, Cc and Dd), would be measured using a laser profilometer. Accordingly, the surface area surrounded by points ABCD could be approximately calculated by using Heron's formula, which is shown as Eq. (6). Obviously, the finer the scale (δ), the more accurate the surface area, and the area may be larger. After varying the value of δ, the different areas of ABCD can be calculated correspondingly. Relationship exists between δ and the surface area AT(δ), shown as Eq. (7). The fractal dimension can be calculated using Eq. (8).
(6)
where Pij=12(|AB|+|BC|+|AC|) and Qij=12(|AD|+|DC|+|AC|)
(7)
where AT(δ)=∑i=1N(δ)Ai(δ)⁠, Dis fractal dimension.
(8)

2.3 Skid resistance test

Skid resistance was measured using a British pendulum tester (Fig. 9). The device is widely used both in laboratory and field tests. The procedure accorded to ASTM E303-2018. The British pendulum number (BPN) could be directly measured from each test. BPN is approximately 100 times the coefficient of friction of a sample. BPN can be regarded as a proxy measure of the microtexture of a pavement to contribute to pavement friction [36]. BPN depends on aggregate types, mix design formula, construction methods and compaction quality [37]. In this study, the test was performed in triplicate.

3. Results and discussion

3.1 Skid resistance

BPNs were obtained using the British pendulum tester. The MTDs of A and B were also tested. Results of surfaces A and B are shown in Fig. 10. It can be observed that BPN of surface A was lower than that of surface B. With the increase of MTD, skid resistance linearly increased accordingly. Generally, MTD is affected by lots of factors, including aggregate gradation, binder content, compaction, aggregate morphology, etc. In this study, the aggregate gradation and bitumen content together led to the difference of MTD. As MTD is an indicator directly representing the macrotexture of a pavement, it indicates the macrostructure significantly affects the skid resistance, which agrees with other studies [16,38]. Asphalt mixtures in A and B have almost the same porosity, while A has a finer gradation than B, indicating a coarse gradation may lead to a larger MTD and a larger skid resistance. It has been reported that the aggregate gradation is a key factor determining the macrotexture, while it has a minor influence on microtexture [8].

3.2 Relationships between roughness parameters and skid resistance

3.2.1 Interface roughness, skewness and kurtosis

Fig. 11 shows the relationships between BPN and Ra, Rsk and Rku. It can be seen that BPN increased along with the increase of Ra, indicating a lager interface roughness value may lead to a more pronounced friction between pavement and tyre. When Ra increased from 0.93 to 1.80 mm, the skid resistance improvement was 21.3%. Rsk values of pavement A and B were lower than 0, indicating both pavement A and B are mainly made up of valleys. It is very easy to understand that after compaction, the pavement surfaces (A and B) contain some pits or voids on the surfaces, and these pits can be regarded as valleys. Rku values of pavement A and B are larger than 3 indicating that surface A and B are both spiky surfaces. However, Rku of pavement B is closer to 3, which indicates the profile height distribution of pavement B is closer to a normal distribution (Fig. 5). Florková and Pepucha [39] measured the micotexture parameters of six sections and correlated these parameters to the friction resistance. The relationships between the parameters (Ra, Rsk and Rku) and BPN were consistent with this study.

Profile height distribution: (a) Rku<3; (b) Rku=3; (b) Rku>3.
Fig. 5.

Profile height distribution: (a) Rku<3; (b) Rku=3; (b) Rku>3.

Open in new tabDownload slide
Sk parameters in material ratio curve.
Fig. 6.

Sk parameters in material ratio curve.

Open in new tabDownload slide
Peak, core and valley in material ratio curve.
Fig. 7.

Peak, core and valley in material ratio curve.

Open in new tabDownload slide
Projective covering method.
Fig. 8.

Projective covering method.

Open in new tabDownload slide
British pendulum tester.
Fig. 9.

British pendulum tester.

Open in new tabDownload slide
BPN vs. MTD.
Fig. 10.

BPN vs. MTD.

Open in new tabDownload slide
BPN vs. (a) Ra; (b) Rsk and (c) Rku.
Fig. 11.

BPN vs. (a) Ra; (b) Rsk and (c) Rku.

Open in new tabDownload slide

3.2.2 Fractal dimension

After obtaining the 3D coordinates and reconstructing the surfaces of A and B, surface areas with different grid sizes can be calculated using Eq. (6). Fig. 12 gives the relationship between grid size and the related surface area, which were derived from Eq. (7). The lower the δ, the finer the grid, and a larger area (AT(δ)) could be obtained. Linear fitting regression was conducted between grid size and the surface area. The fractal dimension (FD) was calculated using Eq. (8). From Fig. 13, it can be seen that FD of surface A was 2.25 and FD of surface B was 2.43. From surface A to B, the skid resistance increased by 21.3%.The closer the value is to 2, the closer the surface is the ideal 2D plane, whose texture depth and roughness (Ra) were 0. FD of pavement A is lower than that of pavement B, and the skid resistance of pavement A is lower accordingly. Some scholars [40,41] concluded that a linear relationship existed between surface roughness and fractal dimension, and a larger FD led to a pronounced surface roughness, thus leading to a significant skid resistance.

Grid size (δ) vs. area (AT(δ)).
Fig. 12.

Grid size (δ) vs. area (AT(δ)).

Open in new tabDownload slide
BPN vs. fractal dimension (FD).
Fig. 13.

BPN vs. fractal dimension (FD).

Open in new tabDownload slide

3.2.3 Material ratio curve

Fig. 14 presents the material ratio curves of pavements A and B. The related parameters are shown in Table 3. The peak cross-sectional area (A1) and valley cross-sectional area (A2) and the related BPN values are shown in Fig. 15. It can be seen that there is no significant difference of A1 between pavements A and B, indicating the peak cross-sectional area may be not an effective factor affecting the skid resistance. Considering the valley cross-sectional area (A2), a larger A2 resulted in a larger BPN, indicating the valley cross-section area may be a positive factor influencing the skid resistance.

Material ratio vs. profile height.
Fig. 14.

Material ratio vs. profile height.

Open in new tabDownload slide
BPN vs. (a) A1 and (b) A2.
Fig. 15.

BPN vs. (a) A1 and (b) A2.

Open in new tabDownload slide
Table 3.
Open in new tab

Parameters related to material ratio curve

ParameterSurface ASurface B
A1 (mm.%)1.061.08
A1_std (mm.%)0.080.11
A2 (mm.%)0.330.64
A2_std (mm.%)0.040.07
Vmp (mm.%)1.061.018
Vmp_std (mm.%)0.090.05
Vmc (mm.%)21.2611.02
Vmc_std (mm.%)2.891.71
Vvc (mm.%)28.714.1
Vvc_std (mm.%)3.102.04
Vvv (mm.%)1.841.11
Vvv_std (mm.%)0.210.15
ParameterSurface ASurface B
A1 (mm.%)1.061.08
A1_std (mm.%)0.080.11
A2 (mm.%)0.330.64
A2_std (mm.%)0.040.07
Vmp (mm.%)1.061.018
Vmp_std (mm.%)0.090.05
Vmc (mm.%)21.2611.02
Vmc_std (mm.%)2.891.71
Vvc (mm.%)28.714.1
Vvc_std (mm.%)3.102.04
Vvv (mm.%)1.841.11
Vvv_std (mm.%)0.210.15
Table 3.
Open in new tab

Parameters related to material ratio curve

ParameterSurface ASurface B
A1 (mm.%)1.061.08
A1_std (mm.%)0.080.11
A2 (mm.%)0.330.64
A2_std (mm.%)0.040.07
Vmp (mm.%)1.061.018
Vmp_std (mm.%)0.090.05
Vmc (mm.%)21.2611.02
Vmc_std (mm.%)2.891.71
Vvc (mm.%)28.714.1
Vvc_std (mm.%)3.102.04
Vvv (mm.%)1.841.11
Vvv_std (mm.%)0.210.15
ParameterSurface ASurface B
A1 (mm.%)1.061.08
A1_std (mm.%)0.080.11
A2 (mm.%)0.330.64
A2_std (mm.%)0.040.07
Vmp (mm.%)1.061.018
Vmp_std (mm.%)0.090.05
Vmc (mm.%)21.2611.02
Vmc_std (mm.%)2.891.71
Vvc (mm.%)28.714.1
Vvc_std (mm.%)3.102.04
Vvv (mm.%)1.841.11
Vvv_std (mm.%)0.210.15

Fig. 16 presents the relationship between Vmp, Vvv and the skid resistance. The values of Vmp were closer to A1, and Vmp can be used to represent the peak area. Similar with A1 in Fig. 15, the difference in Vmp between surfaces A and B is very slight, indicating the main component of pavements A and B is not the peak. Comparing A2 and Vvv, Vvv values were obtained starting from 80% in the material ratio curve. Therefore, Vvv values were significantly larger than A2 values. Considering the volume of dale (Vvv), pavement B obtained a lower Vvv and a larger BPN value.

BPN vs. (a) Vmp and (b) Vvv.
Fig. 16.

BPN vs. (a) Vmp and (b) Vvv.

Open in new tabDownload slide

Fig. 17 presents the relationship between Vmc, Vvc and the skid resistance. Pavement A shows larger Vmc and Vvc values, which was mainly caused by the wider plateau stage in the material ratio curves, leading to a wider profile height range in calculating Vmc and Vvc. For pavement B, the plateau stage is narrow, thus the values of Vmc and Vvc were lower. Lots of studies have revealed that the skid resistance is affected by coupled effects of both macrotexture and microtexture. Considering the parameters in material ratio curve, the coupling effect of these factors rather than just one factor should be also taken into account, since the parameters together make up the curve. If there was another specimen with the same Vmc and Vvc as specimen A, the Vmp and Vvv of the specimen may be different from specimen A, leading to different friction performance. This study just explored the effect of various parameters on the two surfaces, and the coupling effects between these factors will be studied in the future.

BPN vs. (a) Vmc and (b) Vvc.
Fig. 17.

BPN vs. (a) Vmc and (b) Vvc.

Open in new tabDownload slide

4. Conclusions

In this study, some parameters related to the pavement surface morphology were obtained by conducting the sand patching test and 3D laser scanning tests. Skid resistance was obtained using a British pendulum tester. Relationship between the morphology parameters and skid resistance were built. According to the study, conclusions can be drawn as follows.

  1. Mean texture depth (MTD), interface roughness (Ra), skewness (Rsk), kurtosis (Rku), fractal dimension (D) and parameters related to material ratio curve were obtained.

  2. With the increase of the MTD, the skid resistance increased accordingly.

  3. Fractal dimension (FD) can be used to evaluate the skid resistance of a pavement. A larger FD induced a more significant skid resistance.

  4. It is recommended to conduct some analysis about the coupling effect of the parameters related to the material ratio curve in the future. On the other hand, more specimens with different surfaces would be recommended to be fabricated and further analysis should be performed.

ACKNOWLEDGEMENTS

The research was funded by the National Natural Science Foundation of China (Grant Nos. 52008405 and 51778638), to which the authors are very grateful.

Conflict of interest

The authors declare no conflict of interest.

References

1.

Hall
J
,
Smith
K
,
Titus-Glover
L
, et al.
Guide for Pavement Friction; Contractor's Final Report for NCHRP-Project 01-43
.
Transportation Research Board
,
Washington, DC, USA
2009
.

Google Scholar

OpenURL Placeholder Text

 
2.

Wallman
CG
,
Åström
H
.
Friction measurement methods and the correlation between road friction and traffic safety: a literature review
.
Swedish National Road and Transport Research Institute
,
2001
, Project Code is 80435.

3.

Lu
T
.
Study on freeway operational safety management system under ice climate
.
J Highway Transport Res Develop
.
2009
;
1
:
87
–
88
.

Google Scholar

OpenURL Placeholder Text

 
4.

Mayora
JMP
,
Pina
RJ
.
An assessment of the skid resistance effect on traffic safety under wet-pavement conditions
.
Accid Anal Prev
.
2009
;
41
:
881
–
886
.

Google Scholar

Crossref
Search ADS
PubMed

 
5.

Zhang
J
,
Ding
L
,
Zeng
L
, et al.
Using portable falling weight deflectometer to determine treatment depth of subgrades in highway reconstruction of Southern China
.
Transport Saf Envt
.
2020
;
2
:
18
–
28
.

Google Scholar

Crossref
Search ADS

 
6.

Corley-Lay
JB
.
Friction and surface texture characterization of 14 pavement test sections in Greenville, North Carolina
.
Transp Res Rec
.
1998
;
1639
:
155
–
161
.

Google Scholar

Crossref
Search ADS

 
7.

Henry
JJ
.
Evaluation of Pavement Friction Characteristics, NCHRP SYNTHESIS 291
.
Washington D.C., USA
2000
.

Google Scholar

OpenURL Placeholder Text

 
8.

Kogbara
RB
,
Masad
EA
,
Kassem
E
, et al.
A state-of-the-art review of parameters influencing measurement and modeling of skid resistance of asphalt pavements
.
Constr Build Mater
.
2016
;
114
:
602
–
617
.

Google Scholar

Crossref
Search ADS

 
9.

Roe
P
,
Hartshorne
S
.
Polished Stone Value of Aggregates and In-Service Skidding Resistance
.
Thomas Telford
1999
.

Google Scholar

OpenURL Placeholder Text

 
10.

Serigos
PA
,
Smit
ADF
,
Prozzi
JA
.
Incorporating surface microtexture in the prediction of skid resistance of flexible pavements
.
Transp Res Rec
.
2014
;
2457
:
105
–
113
.

Google Scholar

Crossref
Search ADS

 
11.

Cong
L
,
Wang
T
.
Effect of fine aggregate angularity on skid-resistance of asphalt pavement using accelerated pavement testing
.
Constr Build Mater
.
2018
;
168
:
41
–
46
.

Google Scholar

OpenURL Placeholder Text

 
12.

Li
QJ
,
Zhan
Y
,
Yang
G
, et al.
Pavement skid resistance as a function of pavement surface and aggregate texture properties
.
Int J Pavement Eng
.
2020
;
21
:
1159
–
1169
.

Google Scholar

Crossref
Search ADS

 
13.

Hu
L
,
Yun
D
,
Liu
Z
, et al.
Effect of three-dimensional macrotexture characteristics on dynamic frictional coefficient of asphalt pavement surface
.
Constr Build Mater
.
2016
;
126
:
720
–
729
.

Google Scholar

Crossref
Search ADS

 
14.

Din
IMU
,
Mir
MS
.
Effect of copper slag and reclaimed asphalt pavement on the skid resistance of asphalt mixes
.
Int J Pavement Eng
.
2021
;
1
–
14
.

Google Scholar

OpenURL Placeholder Text

 
15.

Alsheyab
MA
,
Khasawneh
MA
.
Quantifying the effect of modified mixture volumetrics and compaction effort on skid resistance of asphalt pavements
.
Int J Pavement Eng
.
2020
;
1
–
9
.

Google Scholar

OpenURL Placeholder Text

 
16.

Plati
C
,
Pomoni
M
.
Impact of traffic volume on pavement macrotexture and skid resistance long-term performance
.
Transp Res Rec
.
2019
;
2673
:
314
–
322
.

Google Scholar

Crossref
Search ADS

 
17.

Chou
CP
,
Lee
CC
,
Chen
AC
, et al.
Using a constructive pavement texture index for skid resistance screening
.
International Journal of Pavement Research and Technology
.
2017
;
10
:
360
–
368
.

Google Scholar

Crossref
Search ADS

 
18.

Kane
M
,
Edmondson
V
.
Long-term skid resistance of asphalt surfacings and aggregates’ mineralogical composition: generalisation to pavements made of different aggregate types
.
Wear
.
2020
;
454–455
:
203339
.

Google Scholar

OpenURL Placeholder Text

 
19.

Fwa
TF
,
Chu
L
.
The concept of pavement skid resistance state
.
Road Materials Pavement Des
.
2021
;
22
:
101
–
120
.

Google Scholar

Crossref
Search ADS

 
20.

Rezaei
A
,
Masad
E
,
Chowdhury
A
.
Development of a model for asphalt pavement skid resistance based on aggregate characteristics and gradation
.
J Transp Eng
.
2011
;
137
:
863
–
873
.

Google Scholar

Crossref
Search ADS

 
21.

Do
M
,
Cerezo
V
,
Beautru
Y
, et al.
Influence of thin water film on skid resistance
.
J Traffic Transport Engng
.
2014
;
2
:
36
–
44
.

Google Scholar

OpenURL Placeholder Text

 
22.

Dan
HC
,
He
LH
,
Xu
B
.
Experimental investigation on skid resistance of asphalt pavement under various slippery conditions
.
Int J Pavement Eng
.
2017
;
18
:
485
–
499
.

Google Scholar

Crossref
Search ADS

 
23.

Tang
T
,
Anupam
K
,
Kasbergen
C
, et al.
A finite element study of rain intensity on skid resistance for permeable asphalt concrete mixes
.
Constr Build Mater
.
2019
;
220
:
464
–
475
.

Google Scholar

Crossref
Search ADS

 
24.

Song
W
,
Huang
B
,
Shu
X
.
Laboratory investigation of fog-seal treatment on performance of open-graded friction course pavement
.
J Mater Civ Eng
.
2021a
;
33
:
04021002
.

Google Scholar

Crossref
Search ADS

 
25.

Song
W
,
Xu
F
,
Wu
H
, et al.
Investigating the skeleton behaviors of open-graded friction course using discrete element method
.
Powder Technol
.
2021b
;
385
:
528
–
536
.

Google Scholar

Crossref
Search ADS

 
26.

Wu
H
,
Yu
J
,
Song
W
, et al.
A critical state-of-the-art review of durability and functionality of open-graded friction course mixtures
.
Constr Build Mater
.
2020
;
237
:
117759
.

Google Scholar

Crossref
Search ADS

 
27.

Wu
H
,
Liu
Z
,
Sun
B
, et al.
Experimental investigation on freeze–thaw durability of Portland cement pervious concrete (PCPC)
.
Constr Build Mater
.
2016
;
117
:
63
–
71
.

Google Scholar

Crossref
Search ADS

 
28.

McDaniel
R
.
Performance and maintenance of OGFC surfaces
.
2010
;
https://engineering.purdue.edu/NCSC/services/2010%20Presentations/Performance%20and%20Maintenance%20of%20OGFC%20Surfaces.pdf
.

29.

Huang
B
,
Shu
X
,
Song
W
.
Maintenance Strategies for Open-Graded Friction Course (OGFC)
.
Tennessee Department of Transportation
,
Nashville, TN, USA
2019
.

Google Scholar

OpenURL Placeholder Text

 
30.

Das
R
,
Mohammad
LN
,
Elseifi
M
, et al.
Development and validation of a model to predict interface bonding between pavement layers
.
Transp Res Rec
.
2018
;
2672
:
22
–
30
.

Google Scholar

Crossref
Search ADS

 
31.

Song
W
,
Shu
X
,
Huang
B
, et al.
Factors affecting shear strength between open-graded friction course and underlying layer
.
Constr Build Mater
.
2015
;
101
:
527
–
535
.

Google Scholar

Crossref
Search ADS

 
32.

Kokkalis
AG
,
Tsohos
GH
,
Panagouli
OK
.
Consideration of fractals potential in pavement skid resistance evaluation
.
J Transp Eng
.
2002
;
128
:
591
.

Google Scholar

Crossref
Search ADS

 
33.

Xie
H
,
Wang
JA
,
Stein
E
.
Direct fractal measurement and multifractal properties of fracture surfaces
.
Phys Lett A
.
1998
;
242
:
41
–
50
.

Google Scholar

Crossref
Search ADS

 
34.

Grassberger
P.
On efficient box counting algorithms
.
Int J Modern Physics C
.
1993
;
4
:
515
–
523
.

Google Scholar

Crossref
Search ADS

 
35.

Chaudhuri
BB
,
Sarkar
N
.
Texture segmentation using fractal dimension
.
IEEE Trans Pattern Anal Mach Intell
.
1995
;
17
:
72
–
77
.

Google Scholar

Crossref
Search ADS

 
36.

Chen
D
,
Han
S
,
Ling
C
, et al.
Prediction of asphalt mixture surface texture level and its distributions using mixture design parameters
.
Int J Pavement Eng
.
2019
;
20
:
557
–
565
.

Google Scholar

Crossref
Search ADS

 
37.

Asi
IM
.
Evaluating skid resistance of different asphalt concrete mixes
.
Building Envt
.
2007
;
42
:
325
–
329
.

Google Scholar

Crossref
Search ADS

 
38.

Torbruegge
S
,
Wies
B
.
Characterization of pavement texture by means of height difference correlation and relation to wet skid resistance
.
J Traffic Transport Engng (Engl Ed)
.
2015
;
2
:
59
–
67
.

Google Scholar

Crossref
Search ADS

 
39.

Florková
Z
,
Pepucha
LU
.
Microtexture diagnostics of asphalt pavement surfaces
,
IOP Conference Series: Materials Science and Engineering
.
IOP Publishing
,
2017
; p.
012025
.

Google Scholar

Crossref
Search ADS

 
40.

Cao
Rh
,
Wang
C
,
Yao
R
, et al.
Effects of cyclic freeze-thaw treatments on the fracture characteristics of sandstone under different fracture modes: laboratory testing
.
Theor Appl Fract Mech
.
2020
;
109
:
102738
.

Google Scholar

Crossref
Search ADS

 
41.

Tang
Z
,
Huang
F
,
Peng
H
.
Effect of 3D roughness characteristics on bonding behaviors between concrete substrate and asphalt overlay
.
Constr Build Mater
.
2021
;
270
:
121386
.

Google Scholar

Crossref
Search ADS

 
© The Author(s) 2022. Published by Oxford University Press on behalf of Central South University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
Download all slides