Road Materials and Pavement Design

ISSN: 1468-0629 (Print) 2164-7402 (Online) Journal homepage: http://www.tandfonline.com/loi/trmp20

Evaluation method of pavement surface

characteristics for rolling resistance

Atsushi Kawakami, Tsutomu Ishigaki, Yu Shirai, Masaru Terada & Kazuyuki

Kubo

To cite this article: Atsushi Kawakami, Tsutomu Ishigaki, Yu Shirai, Masaru Terada & Kazuyuki
Kubo (2017): Evaluation method of pavement surface characteristics for rolling resistance, Road
Materials and Pavement Design, DOI: 10.1080/14680629.2017.1304264

To link to this article: http://dx.doi.org/10.1080/14680629.2017.1304264

Published online: 27 Mar 2017.

Submit your article to this journal

Article views: 16

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=trmp20

Download by: [The UC San Diego Library] Date: 10 April 2017, At: 05:32
Road Materials and Pavement Design, 2017
http://dx.doi.org/10.1080/14680629.2017.1304264

Evaluation method of pavement surface characteristics for rolling

resistance
Atsushi Kawakamia∗ , Tsutomu Ishigakib , Yu Shiraib , Masaru Teradaa and Kazuyuki Kuboa
a Public Works Research Institute (PWRI), Tsukuba-shi, Japan b NIPPO Corporation, Saitama-shi, Japan

(Received 12 October 2016; accepted 15 February 2017 )

Pavement technology intended to reduce tyre-pavement rolling resistance is a new technol-

ogy for establishing the future low-carbon society. Reduction of rolling resistance leads to
reduce vehicle CO2 emission and to improve fuel efficiency. However, the relation between
rolling resistance and pavement surface characteristics is complex and there are some issues
to be solved including clarifying the evaluation methods. The evaluation method for rolling
resistance generally involves measuring the rolling resistance coefficient directly; however
this direct measurement requires special measurement devices. Indirect evaluation methods
will make measurements easier and help put low-rolling resistance pavement technologies to
practical use. This paper reports the results on determining the contact pressure distribution
between tyre and pavement surfaces using a pressure measurement film to evaluate the rolling
resistance indirectly. The results of analyses the relations between the rolling resistance and
pavement surface characteristics, it were revealed that the contour map of contact pressure
distribution were fully different depending on the pavement surface texture and it would be
useful to evaluate the negative texture that lead to the low-rolling resistance pavement surface.
Keywords: pavement surface; contact pressure distribution; rolling resistance; negative
texture

1. Introduction
On 17 July 2015, at the meeting of the Global Warming Prevention Headquarters, the Japanese
government set a target to reduce greenhouse gas emissions by 26% from 2013 levels by 2030
(Global Warming Prevention Headquarters, 2015). In 2013, Japan’s greenhouse gas emissions
totalled 1408 million tons in terms of the carbon dioxide (CO2 ) equivalent, 11.7% above 1990
reference levels. This increase has led to calls for the development of new technologies that
reduce emissions dramatically.
Pavement technology intended to reduce tyre-pavement rolling resistance (‘rolling resistance’
hereinafter) is a new technology that has the possibility to establish a low-carbon society because
this technology also reduces vehicle CO2 emissions and improves fuel efficiency. Some research
reports related to the rolling resistance have been published mainly from the European countries
and North America (Bendtsen, 2004; Sandberg, 2011; Willis, Robbins, & Thompson, 2014). As
a major precedent study case, the road surface characteristics influence the rolling resistance
and fuel consumption (Sandberg, 1990). The macrotexture and megatexture influence the rolling
resistance (Descornet, 1990). Rolling resistance was estimated by IRI (International Roughness
Index) and MPD (Mean Profile Depth) in PIARC (World Road Association) HDM-4 (Bennett

*Corresponding author. Email: kawakami@pwri.go.jp

© 2017 Informa UK Limited, trading as Taylor & Francis Group

2 A. Kawakami et al.

Table 1. Performance indices and evaluation methods for low-rolling resistance pavement technologies.

Measurement methods

Pavement Performance Direct measurement Indirect measurement

performance index methods methods

(1) Reduction in Rolling resistance Measuring rolling Calculation of rolling

rolling resistance coefficient resistance using a skid resistance coefficient
resistance measurement by surface texture
system characteristics
(2) Improvements Vehicle fuel Fuel gauge measurements Calculations based on mode
in fuel efficiency efficiency fuel economy
(3) Reductions in Vehicle CO2 Measuring CO2 emission Calculations based on CO2
CO2 emissions emissions using an exhaust gas emissions intensity
measurement system

& Greenwood, 2001). Authors have also developed the pavement technology of the low-rolling
resistance and researched the mechanisms for reducing rolling resistance according to the con-
ditions of pavement in Japan (Ishigaki, Shirai, Kawakami, & Kubo, 2015a; Kawakami, Terada,
Kubo, & Ishigaki, 2014; Shirai, Ishigaki, Kawakami, & Kubo, 2015).
This paper reports the results of the study on determining the contact pressure distribution
between tyre and pavement surfaces using a pressure measurement film to evaluate the rolling
resistance. This study is related to joint research between the Public Works Research Institute
(PWRI) and NIPPO Corporation on improving low-carbon pavement technologies.

2. Procedures
Vehicle CO2 emissions and fuel efficiency depend heavily on resistance to vehicle motion. Resis-
tance to vehicle motion generally falls into one of four categories: aerodynamic resistance, rolling
resistance, gradient resistance and acceleration resistance (Sakai, 2001). Rolling resistance is the
factor directly affected by pavement surface characteristics. Reducing the rolling resistance of
pavement surfaces can help reduce vehicle CO2 emissions and enhance fuel efficiency. Table 1
shows evaluation methods for pavement rolling resistance (Ishigaki, Shirai, Kawakami, & Kubo,
2015b). The pavement evaluation methods for a low-rolling resistance pavement can be classi-
fied into three groups based on pavement performance: (1) reductions in rolling resistance; (2)
improvements in fuel efficiency; and (3) reductions in CO2 emissions. The measurement methods
can also be divided into two groups, direct or indirect, based on the quantitative evaluation of the
performance indices. The evaluation method for rolling resistance generally involves measur-
ing the rolling resistance coefficient directly; however direct measurements of rolling resistance
coefficient require special measurement devices. Indirect evaluations will make measurements
easier and help put low-rolling resistance pavement technologies to practical use. In this paper,
the distribution of contact pressure between tyre and pavement surface was focused on as one of
the indirect evaluation methods. The paper explores its characteristics and discusses the potential
for indirect evaluation.

2.1. Measuring the distribution of contact pressure between tyre and pavement surface
To explore the indirect evaluation method for rolling resistance, the contact pressures between
tyre and pavement surfaces were measured. A vehicle equipped with a skid resistance mea-
surement system was used for measurement, and the tyre for skid resistance measurements was
 Road Materials and Pavement Design 3

Figure 1. Measuring distribution of contact pressure between tyre and pavement surface. (a) Tyre 5 of
skid resistance measurement vehicle; (b) a closeup during measurement.

Table 2. Testing conditions of contact pressure distribution.

Item Conditions
Type of test tyre 165 S R13 (rubber content meets ASTME500)
Load of test tyre 3969 N ± 98 N
Tyre inflation pressure 0.176 MPa
Loading time 2 minutes (apply load for about 5 seconds until
the load reaches the specified level)
Pressure measurement film (pressure Film 1; 0.05–0.2 MPa
range) Film 2; 0.2–0.6 MPa
Film 3; 0.5–2.5 MPa
Number of measurement locations 2 locations/test section

statically loaded on a pressure measurement film for the specified duration (Figure 1). Table 2
gives the measurement conditions. The vertical load applied to the test tyre was 3969 ± 98 N,
same as the skid resistance measurement in Japan. Three types of pressure measurement films
were used, and each type was designed for a specific pressure range. Applying pressure to a
measurement film breaks microcapsules inside the film, allowing the colour-forming material in
the microcapsules to react with the colour-developing material and turning the areas red. After
measuring distribution of contact pressure between tyre and pavement surface, these films were
read using a special scanner, performed a pressure analysis of the colour maps, and calculated
the contact area for each pressure.
For testing, twenty-two test sections of pavements at the test course of the National Insti-
tute for Land and Infrastructure Management (NILIM) (Figure 2) were used. Table 3 shows
the main characteristics of the pavement surfaces. Pavement types include dense graded (DG),
porous asphalt (PA), stone mastic asphalt (SMA), continuously reinforced concrete (CO) and
low-rolling resistance asphalt pavements (LRRAPs). LRRAPs were developed by PWRI and
NIPPO Corporation in the collaboration research from 2009 through 2015 (Ishigaki et al.,
2015a). LRRAPs’ mixture consists of aggregate, the maximum size of which is 5 mm, filler
and specially treated polymer-modified asphalt binder. The textures of LRRAPs are formed
to a negative texture by devising aggregate gradation and construction method. The mea-
surement courses of LRRAPs 1–4 used in this test were constructed as the prototype in the
development stage.
Pavement surface characteristics were measured by texture evaluation methods, such as
MPD, SMTD (Sensor Measured Texture Depth), MTD (Mean Texture Depth) and BPN (British
Pendulum Number). These measurements were conducted in accordance with the pavement
investigation and examination method which was authorised by the Japan Road Association
(2007). These pavements of test sections varied in aggregate particle size and pavement surface
degradation.
4 A. Kawakami et al.

Figure 2. Twenty-two test sections of pavements at the test course of NILIM.

Table 3. Test sections and pavement surface characteristics.

Pavement (stone
No. max. size)* Length (m) MPD (mm) SMTD (mm) MTD (mm) BPN

1 PA(13) 60 1.54 1.37 2.01 67

2 DG(20) 60 0.35 0.32 0.30 69
3 LRRAPs 1 60 0.61 0.46 0.67 69
4 LRRAPs 2 60 0.58 0.41 0.56 64
5 LRRAPs 3 60 0.71 0.50 0.96 62
6 LRRAPs 4 60 0.62 0.42 0.64 63
7 DG Cr 95 0.79 0.47 0.92 64
8 CO Cr 80 0.44 0.29 0.47 66
9 SMA(5) 120 0.72 0.51 0.71 66
10 SMA(13) 120 1.21 1.08 1.26 69
11 PA(5) Cr 120 1.01 0.69 1.16 55
12 PA(5) 80 1.07 0.68 1.37 55
13 PA(13) 80 1.79 1.26 1.80 47
14 DG 80 0.71 0.34 0.65 61
15 CO Cr 80 0.40 0.21 0.44 66
16 PA(20) 55 3.50 2.67 4.80 62
17 PA(20) 45 2.45 1.80 2.81 64
18 PA(10) 55 1.49 0.93 1.36 54
19 PA(13) 55 1.93 1.21 1.89 51
20 PA(13) 60 2.11 1.49 2.05 56
21 DG Cr 95 0.84 0.46 0.81 58
22 CO 80 0.52 0.23 0.48 63
*There are some longitudinal cracks on the pavement surface.

2.2. Direct measurement of rolling resistance coefficient

Rolling resistance is the traction force applied by a road surface that affects a tyre in the direction
opposite to the travelling direction. It is generated between the rolling tyre and the contact area of
the pavement surface. In this measurement, the traction resistance of the test tyre was measured
using a vertical-lift longitudinal and transverse skid resistance measurement system. Table 4
shows the measurement conditions. The survey vehicle speeds were constant at 20, 40, 60 and
80 Km/h. The numbers of measurement were each 10 times.
 Road Materials and Pavement Design 5

Table 4. Overview of skid resistance measurement system.

Item Conditions

Measuring device Vertical-lift longitudinal and transverse skid resistance measurement

system
Measurement item Traction resistance (Rc), Cornering force (Fc), Vertical load (Fz),
Velocity of test tyre (V), Temperature of test tyre (T), Data sampling
interval; 0.01 (s)
Type of test tyre 165 S R13 (rubber content meets ASTME500)
Load of test tyre 4000 N ± 150 N
Tyre inflation pressure 0.176 MPa
Velocity 20, 40, 60, 80 (Km/h)
Number of measurement 10 times at each speed/each section

2.3. Exploring an indirect method for evaluating rolling resistance

Based on the contact area for each pressure, the characteristics of contact pressure distribution
and their relations to pavement surface characteristics were analysed. Also the relations with
measured rolling resistance were analysed to explore the possibility of using the distribution of
contact pressure between tyre and pavement surface as an indirect evaluation method for the
rolling resistance.

3. Result of the study

3.1. Distribution of contact pressure between tyres and pavements on the test road surface
Figure 3 shows some of the contour maps of the contact pressure distribution obtained in this
study. It shows that the distribution markedly varies with the pavement texture. For example, in
section 4, all contact areas are large, and contact pressure is uniform. In section 16, all the contact
areas are small and the contact pressures are locally high.
By visible observation, section 4 can be roughly categorised as a negative-textured pavement
surface having uniform texture, while section 16 can be categorised as a positive-textured pave-
ment surface having coarse texture. Figure 4 shows contact areas in each test section ordered by
area size. The measurement of contact area was conducted at two locations in each section, the
variability between two measurements of contact area in each test section was less than ± 5.2%
from the average. Figure 4 suggests that the contact areas are large in test sections with a dense
texture, such as DG pavement, CO pavement and LRRAP. The contact areas are small in sec-
tions of coarse texture, like the PA pavement. However, it is difficult to distinguish between the
negative-textured surface and positive-textured surface using only this contact area.
Figure 5 shows changes in the contact area for each contact pressure. To the changes in the
contact area for each contact pressure, the contact area on a flat surface is added for comparison.
If contact pressures are less than 0.50–0.75 MPa, the contact areas on the test pavements are
smaller than on the flat surface. If the contact pressures exceed this range, the contact areas are
larger. Therefore, the contact pressures at 0.50, 0.75 and 1.75 MPa were decided to divide and
add each pressure area for quantitative analysis.

3.2. Exploration of evaluation methods of rolling resistance

It is assumed that the area with low contact pressures is greater than the area with high contact
pressures on a negative-textured pavement surface and that this is reversed for the positive-
textured pavement surface. For each pavement surface, the ratio of areas with low contact
6 A. Kawakami et al.

Figure 3. Contour maps of the contact pressure distribution between tyre and pavement surface (excerpt).

Figure 4. Contact areas in each test section.

pressures (0.5 MPa or less) and those with high contact pressures (1.75 MPa or more) were
analysed. As Figure 6 shows, they are closely correlated. Therefore, it is considered that the
tendency of the contact pressure distribution can be estimated by the measuring the contact area
and analysing the ratio of the area with high contact pressure. Figure 7 shows the ratios of the
 Road Materials and Pavement Design 7

Figure 5. Changes in area for each contact pressure (test pavements and flat surface).

Figure 6. Relations between ratio of the area with contact pressure at 0.5 MPa or less and ratio of area at
1.75 MPa or more.

areas with high contact pressures arranged in ascending order. It was believed that contact pres-
sure of an uneven textured surface tends to be greater locally. That is why the ratio of the area
with high contact pressures would be low on a negative-textured pavement but high on a positive-
textured pavement. The result suggests that pavement surfaces could be quantitatively classified
into negative- and positive-textured surfaces.
Table 5 shows correlations for the existing texture evaluation methods, such as MPD, SMTD,
MTD and BPN. In the MPD, SMTD and MTD used for texture roughness, these indices appar-
ently correlate negatively with the ratio of low contact pressure area and correlate positively
with the ratio of high contact pressure area. Moreover, the correlation coefficient between the
ratio contact pressure area and texture evaluation index was the highest in the ratio of area with
contact pressure at 1.75 MPa. In BPN, an index for skid resistance, it does not highly correlate
with the ratio of the contact pressure area. It was found that the distribution of contact pres-
sure between tyre and pavement surface would be useful to quantitatively classify into negative
texture and positive texture that lead to the low-rolling resistance pavement surface, just as for
existing texture evaluation indices. By using the contact pressure distribution or combining it
8 A. Kawakami et al.

Figure 7. Order of ratios of high contact pressures (1.75 MPa or more) area.

Table 5. Relations in existing texture evaluation indices.

Ratio of contact pressure area

Correlation coefficient
Index 0.05–0.50 0.50–0.75 0.75–1.75 1.75–
MPD* − 0.84 − 0.11 0.77 0.89
SMTD* 0.77 0.02 0.70 0.80
MTD* − 0.85 − 0.04 0.79 0.87
BPN 0.52 0.14 − 0.53 − 0.53
*Correlation coefficient obtained by logarithmic conversion of data and linear approximation.

with an existing texture evaluation method, negative- and positive-textured surfaces could be
divided and a low-rolling resistance pavement could be indirectly evaluated.

3.3. Comparison to rolling resistance

Figure 8 shows measurement results for the rolling resistance of each pavement surface. The
test sections on the horizontal axis in this figure are arranged, from lowest to highest, by ratio
of high contact pressure (1.75 MPa or more) area, same as in Figure 7. At all vehicle speeds
(20, 40, 60 and 80 Km/h), the rolling resistance coefficient tends to be higher with higher ratios
for the high contact pressure area. Figure 9 shows the measurement result at 20 Km/h for the
correlation between the ratio of the area and the rolling resistance coefficient. As a result, the
correlation coefficients are 0.77, 0.52, 0.09 and 0.19 at vehicle speeds of 20, 40, 60 and 80 Km/h,
respectively. Although rolling resistance data require corrections for certain factors, such as tem-
perature, lateral force and speed, the results of this study show that the distribution of contact
pressure between tyre and pavement surface correlates with the rolling resistance coefficient at
low speed. To make it possible to predict rolling resistance coefficients through indirect evalu-
ations, pavement surface roughness (σ or IRI) and vehicle speed, as well as pavement surface
texture must be considered with the contact pressure distribution used in this study. Additionally,
 Road Materials and Pavement Design 9

Figure 8. Rolling resistance coefficient for each pavement type (rolling resistance coefficient (without
correction for lateral force)).

Figure 9. Relations between rolling resistance coefficient and ratio of the area with contact pressure of
1.75 MPa or greater.

measurement of the distribution of contact pressure between tyre and pavement surface must be
simplified.

4. Conclusion
(1) The distribution of contact pressure between tyre and pavement surface markedly varies
with the pavement texture. By visible observation, these distributions can be roughly
categorised as a negative-textured pavement having uniform texture and as a positive-
textured pavement having coarse texture.
10 A. Kawakami et al.

(2) The ratio of the area with high contact pressures would be low on a negative-textured
pavement but high on a positive-textured pavement. Pavement surfaces would be useful
to quantitatively classify into negative- and positive-textured surfaces that lead to the
low-rolling resistance pavement surface.
(3) In the MPD, SMTD and MTD used for texture roughness, these indices apparently corre-
late negatively with low contact pressure area and correlate positively with high contact
pressure area. In BPN, an index for skid resistance, it does not highly correlate with the
contact area.
(4) The distribution of contact pressure between tyre and pavement surface correlates with
the rolling resistance coefficient at low vehicle speed. These results suggest that the dis-
tribution of contact pressure method with some improvement could be used in indirect
evaluations of rolling resistance.

Disclosure statement
No potential conflict of interest was reported by the authors.

References
Bendtsen, H. (2004). Rolling resistance, fuel consumption – A literature review (Technical note 23). Danish
Road Institute.
Bennett, C., & Greenwood, I. (2001). Modelling road user and environmental effects in HDM-4 (Vol. 7,
The highway development and management series).
Descornet, G. (1990). Road-surface influence on tire rolling resistance. In W. E. Meyer & J. Reichert (Eds.),
Surface characteristics of roadways: International research and technologies, ASTM STP 1031 (pp.
401–415). Philadelphia, PA: American Society for Testing and Materials.
Global Warming Prevention Headquarters. (2015). Submission of Japan’s intended nationally determined
contribution. Tokyo. Retrieved from http://www.mofa.go.jp/files/000090898.pdf
Ishigaki, T., Shirai, Y., Kawakami, A., & Kubo, K. (2015a). Low rolling resistance asphalt pavements for
traffic energy saving and CO2 emission reduction. Seoul: PIARC.
Ishigaki, T., Shirai, Y., Kawakami, A., & Kubo, K. (2015b). Performance indices and evaluation methods
for fuel-efficient pavement (31st Japan Road Conference, no. 3P27).
Japan Road Association. (2007). Handbook of investigation and testing of pavement. Tokyo: Japan Road
Association.
Kawakami, A., Terada, M., Kubo, K., & Ishigaki, T. (2014). Study on the pavement surface properties
and tire rolling resistance, sustainability, In M. Losa, & T. Papagiannakis (Eds.), Eco-efficiency and
conservation in transportation infrastructure asset management (pp. 535–539). London: CRC Press.
Sakai, H. (2001). Tire engineering. Tokyo: Grand Prix Book.
Sandberg, U. (1990). Road macro- and megatexture influence on fuel consumption. In W. E. Meyer & J.
Reichert (Eds.), Surface characteristics of roadways: International research and technologies, ASTM
STP 1031 (pp. 460–479). Philadelphia, PA: American Society for Testing and Materials.
Sandberg, U. (2011). Road surface influence on tyre/road rolling resistance (Report MIRIAM_SP1_04).
Shirai, Y., Ishigaki, T., Kawakami, A., & Kubo, K. (2015). Evaluation methods for tire/pavement rolling
resistance (31st Japan Road Conference, no. 3049).
Willis, J., Robbins, M., & Thompson, M. (2014). Effect of pavement properties on vehicular rolling resis-
tance: A literature review (NCAT report 14-07). Alabama: National Center for Asphalt Technology.