• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.5D
• /
• pp.797-804
• /
• 2006

A development of regression model for asphalt concrete pavements using Falling Weight Deflectometer deflections is presented in this paper. A backcalculation program based on layered elastic theory was used to generate the synthetic modulus database, which was used to generate 95% confidence intervals of modulus in each layer. Using deflection basins of FWD data used in developing this procedure were collected from Pavement Management System in flexible pavements. Assumptions of back-calculation are that one is 3 layered flexible pavement structure and another is depth to bedrock is finite. It is found that difference of between 95% confidence intervals and modulus ranges of other papers does not exist. So, the data of 95% confidence intervals in each layer was used to develop multiple regression models. Multiple regression equations of each layer were established by SPSS, package of Statics analysis. These models were proved by regression diagnostics, which include case analysis, multi-collinearity analysis, influence diagnostics and analysis of variance. And these models have higher degree of coefficient of determination than 0.75. So this models were applied to predict modulus of domestic asphalt concrete pavement at FWD field test.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.14 no.3
• /
• pp.495-508
• /
• 1994

A backcalculation procedure to determine the layer moduli of flexible pavement structure is developed using matrix solution technique. Forward calculation computer program adopted in this backcalculation procedure is SINELA which is one of layered elastic computer programs. Data base system is used as a tool for setting initial seed moduli and depth to virtual bedrock in backcalculation procedure. The validity and applicability of the proposed backcalculation procedure are verified through various numerical model tests. From the results of comparison analysis with FPEDD1, it is found that the proposed procedure gives more efficient and accurate results.

• Journal of the Korean Geosynthetics Society
• /
• v.9 no.3
• /
• pp.57-65
• /
• 2010

The main objective of this study is to investigate the time-dependent deformation of EPS blocks under repeated loading conditions which is the one of the flexible pavement structure. The study comprised of both the experimental work and analytical modeling in order to understand the behavior of EPS blocks under repeated loading. The analytical modeling included the selection of a suitable model for describing the deformation behavior observed under repeated loading conditions, investigating the relationship among the unit weight, deformation and applied stress, analyzing the effect of repeated load on deformation. The test results were compared with the Findley's theory and model analysis with the results of this research under repeated loading conditions. Both Modified Findley's model and the proposed model can be adopted to illustrate the deformation behavior of EPS blocks under repeated loads.

• International Journal of Highway Engineering
• /
• v.19 no.4
• /
• pp.1-7
• /
• 2017

PURPOSES : This study aims to improve complex modeling of multivariable, nonlinear, and overdispersion data with an artificial neural network that has been a problem in the civil and transport sectors. METHODS: Deep learning, which is a technique employing artificial neural networks, was applied for developing a large bus fuel consumption model as a case study. Estimation characteristics and accuracy were compared with the results of conventional multiple regression modeling. RESULTS : The deep learning model remarkably improved estimation accuracy of regression modeling, from R-sq. 18.76% to 72.22%. In addition, it was very flexible in reflecting large variance and complex relationships between dependent and independent variables. CONCLUSIONS : Deep learning could be a new alternative that solves general problems inherent in conventional statistical methods and it is highly promising in planning and optimizing issues in the civil and transport sectors. Extended applications to other fields, such as pavement management, structure safety, operation of intelligent transport systems, and traffic noise estimation are highly recommended.

• Journal of the Korean Geosynthetics Society
• /
• v.9 no.3
• /
• pp.1-8
• /
• 2010

The track systems installed in Korea railway consist of two types on ballasted track or ballastless track. In this study, it was compared with difference of the behaviors at roadbed with cyclic loading through full scale model test. From the results of model tests, loading distribution ratio of the concrete slab track become more widely distributed than ballasted track, and loading distribution ratio at concrete track was about 30:20:15. The concrete slab track is likely to behavior of the rigid plate, while ballasted track is such as flexible pavement. The vertical stresses of upper roadbed with traffic cyclic loading in concrete track were measured about 30 kPa or less. It was a scene very similar to the results of the field train running test. The vertical stress at concrete track was occurred approximately 4 times smaller than ballasted track. Also, the soil velocities with cyclic loading at the slab track were occurred about 0.3 cm/sec or less, its 8 times smaller than ballasted track.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.2D
• /
• pp.227-234
• /
• 2008

This study examines which models are more suitable for representing mechanical property of unbound materials to analyze behavior of asphalt pavement structure. Results from FWD (Falling Weight Deflectometer) test were used to apply to nonlinear elastic model. The new method which can deduct material constants of nonlinear elastic model is suggested from FWD test data rather than laboratory resilient modulus ($M_R$) test. It is confirmed that the material constants are within the common range in subbase. Test output from FWD and MDD (Multi-Depth Deflectometer) was used to verify reliability of the model. From the results of verification, this study shows that a non-linear elastic model agrees to MDD test data more than a linear elastic model does.