• 한국도로학회:학술대회논문집
• /
• 1999.11a
• /
• pp.89-96
• /
• 1999

• 한국도로학회:학술대회논문집
• /
• 1999.11a
• /
• pp.107-110
• /
• 1999

• 한국도로학회:학술대회논문집
• /
• 1999.11a
• /
• pp.111-115
• /
• 1999

• 한국도로학회:학술대회논문집
• /
• 2008.10a
• /
• pp.387-393
• /
• 2008

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.23 no.6
• /
• pp.38-44
• /
• 2019

A Convolution Neural Network(CNN) model was utilized to detect surface cracks in asphalt concrete pavements. The CNN used for this study consists of five layers with 3×3 convolution filter and 2×2 pooling kernel. Pavement surface crack images collected by automated road surveying equipment was used for the training and testing of the CNN. The performance of the CNN was evaluated using the accuracy, precision, recall, missing rate, and over rate of the surface crack detection. The CNN trained with the largest amount of data shows more than 96.6% of the accuracy, precision, and recall as well as less than 3.4% of the missing rate and the over rate.

• International Journal of Highway Engineering
• /
• v.11 no.2
• /
• pp.37-44
• /
• 2009

Currently the reasonability of threshold values for rutting and cracking does not clearly defined at the Pavement Rehabilitation Decision Tree on national highway PMS(Pavement Management System). The goal of this study is to provide the reasonable threshold values for the national highway asphalt concrete pavement rehabilitation. To achieve this goal, test section that represents typical asphalt concrete pavement of national highway was selected and pavement export were participated. Pavement condition survey has been conducted and pavement performance data at the selected roadway section were analyzed. From this study, reasonable threshold values of Pavement Rehabilitation Decision Tree were suggested based on the pavement expert's engineering judgement. In terms of crack repairs, the application of overlay after cutting is required to deteriorated area where existing crack ratio is over 35% and just overlay is required to where crack ratio is over 20%. On rutting, rut depth over 13mm is required to overlay after cutting and rut depth over 10mm is just needed to overlay.

• International Journal of Highway Engineering
• /
• v.11 no.2
• /
• pp.167-174
• /
• 2009

The simulation of dynamic load was conducted based on surface profile on asphalt concrete pavement, vehicle speeds, and suspension types using a truck simulation program. The results of the simulated dynamic load based on different surface profile, vehicle speeds, and suspension types are analyzed. As pavement roughness and vehicle speed are increased, the dynamic load was increased. Walking beam suspension produces greater dynamic load than air spring suspension. Pavement damage index is calculated based on covariance of dynamic load and Paris-Erdogan fracture parameter, n which is based on creep compliance tests of asphalt mixtures used in Korea. The higher covariance of dynamic load, confidence level, and fracture parameter are used, the greater pavement damage index is obtained. Specification of pavement roughness can be developed in various vehicle speeds and asphalt mixtures, and pay factor can be determined after constructing asphalt concrete pavement using pavement damage concepts.

• International Journal of Highway Engineering
• /
• v.5 no.3 s.17
• /
• pp.1-9
• /
• 2003

This study was performed to evaluate the effect of various joint sealant on reducing the reflection cracking of asphalt concrete overlay on cement concrete pavement. The test method used was an accelerated reflection cracking test in shear mode, which was developed for evaluation of reflection cracking resistance of overlaid asphalt concrete in laboratory. The test results showed that use of joint sealants resulted in a significant reduction of reflection cracking. When sealant E was used, the fatigue life was the highest, with relatively larger horizontal deformation. When Sealant B was used, the dynamic stability was the highest with the smallest horizontal deformation. In general, the greater the tensile strength of sealant, the better the crack resistance of the mixture.

• International Journal of Highway Engineering
• /
• v.12 no.4
• /
• pp.111-120
• /
• 2010

This study evaluated strength properties of recycled asphalt concretes using warm-mix technology. Granite with maximum size of 13mm and penetration grade of 80-100 virgin binder were used for mixing in recycled mixtures. Mix design was performed using 20% and 30% RAP(coarse : fine= 6 : 4) contents. GPC, penetration, absolute viscosity and kinematic viscosity were measured for determining ratio of two warm-mix additives (Evotherm and Sasobit). Low-density polyethylene(LD) used as asphalt modifier for improving stiffness of recycled WMA mixtures in this study. Therefore, a total of 11 mixtures were prepared in this study; 8 warm-mix recycled mixtures(2 RAP contents${\times}$2 warm-mix additives${\times}$2 modifiers), 2 hot-mix recycled mixtures and 1 HMA virgin mixture(control). Deformation strength, indirect tensile strength, moisture sensitivity, permanent deformation by wheel tracking tests were measured out for evaluating fundamental properties of recycled asphalt concretes using warm-mix technology.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.6D
• /
• pp.599-605
• /
• 2012

It is important to decide traffic opening time for construction plan of epoxy asphalt pavement. For this purpose, strength prediction model of epoxy asphalt concrete is required. In this study, Marshall stability was measured according to temperature and time for making strength properties equation. Strength prediction model was developed using chemical kinetics considering temperature variation. The traffic opening time of epoxy asphalt pavement on bridge deck has been predicted using the developed model. The prediction and actual traffic opening times were different by 17-days, because weathers of year 2009-2011 used in prediction model were different from weather of year 2012. When the prediction model used the actually measured temperatures of pavement, the difference between real opening time and prediction opening time was two days. The correlation analysis result between measured strength and prediction strength revealed that the $R^2$ using accurate temperature of pavement was 0.95. An improved precise prediction result is to be obtained if the prediction model uses accurate temperature data of pavement.