Serdal Terzi

Modeling deflection basin using artificial neural networks with cross-validation technique in backcalculating flexible pavement layer moduli

Through the new technological developments, for highway maintenance engineering the structural capacity of pavement is to be determined using non-destructive techniques. Up to now various methodologies have been applied based on the surface deflection bowl obtained under either a known moving wheel load or devices such as falling weight deflectometer. Backcalculating pavement layer moduli are well-accepted procedures in the evaluation of the structural capacity of pavements. The ultimate aim of the backcalculation process from non-destructive testing (NDT) results is to estimate the pavement material properties. Using backcalculation analysis, in situ material properties can be backcalculated by the measured field data for appropriate analysis techniques. To backcalculate reliable moduli, the deflection basin must be modeled more realistically. Here, in this study, the deflection basins measured on the surface of the flexible pavements are modeled using artificial neural networks (ANN) with cross-validation technique. Distances between transducers can be varied with different producer companies. The distances between transducer are used for the form deflection basin. Layer thickness and distance to loading center are used as input in the present study. Limited experimental deflection data groups from NDT are used to show the capability of the neural network technique in modeling the deflection bowl. Since enough data are not available to construct a reliable neural network, a methodology based on the cross-validation technique can be used. The results show that the proposed methodology give the deflection bowl satisfied accuracy.

Backcalculation of pavement layer moduli and Poisson's ratio using data mining

Pavement deflection data are often used to evaluate a pavements structural condition non-destructively. Pavement layers are characterized by their elastic moduli estimated from surface deflections through backcalculation. Using backcalculation analysis, flexible pavement layer in situ material properties can be backcalculated from the measured field data through appropriate analysis techniques. This study focuses on the use of data mining (DM)-based pavement backcalculation tools for determining the in situ elastic moduli and Poissons ratio of asphalt pavement from synthetically derived Falling Weight Deflectometer (FWD) deflections at seven equidistant points. In estimation of the elastic modulus and Poissons ratio, data mining (DM) method has not been used as a backcalculation tool before. Experimental deflection data groups from NDT are used to show the capability of the DM approaches in backcalculating the pavement layer thickness and compared each other. By looking at the results of the study, Kstar method gives fine results with respect to other DM methods. Backcalculation of pavement layer elastic modulus and Poissons ratio with DM has been carried out for the first time.

Planning maintenance works on pavements through ant colony optimization

Pavements constructed for the purpose of meeting the demand of highways which were emerged with the improving technological developments increased. And consequently, more resources were demanded to be directed to pavement maintenance and rehabilitation. Hereby, the concept of pavement management emerged. Although project-level analyses were found adequate previously, network-level evaluations were needed in order to do detailed planning as a result of resource allocation and transfer problems that were emerged later. Therefore, pavement management system has become compulsory for all pavements to be controlled together. In this framework, programming is needed in order to schedule maintenance–rehabilitation and develop costs with respect to budget. In the study carried out, a mode was developed in order to program the routine network maintenance activities in terms of Pavement Maintenance and Management Systems, and it was concluded that this problem can be solved through ant colony, using Visual Basic.

Performance model for asphalt concrete pavement based on the fuzzy logic approach

AbstractAccurate pavement performance estimation is very important for the managing and maintaining of surface transportation infrastructure. In the present study, a new model for the prediction of present and future performance of flexible pavements is developed using the fuzzy logic approach. The database of the model is able to use numerical measurements and also linguistic statements. Many models developed in the literature neglect the parameters that have little bearing on performance. However, it is a well known fact that these parameters do have an effect on performance to some degree. Different parameters were considered in the model as accepted by the authorities, and as having little bearing on performance. For each parameter, a certain weight was appointed, and the parameters that affected performance were assigned greater weights while the others were assigned smaller weights. As a result, the current model in the fuzzy logic approach is more flexible than the current Pavement Serviceability I...

Optimization of the deflection basin by genetic algorithm and neural network approach

This paper introduces a new concept of integrating artificial neural networks (ANN) and genetic algorithms (GA) in modeling the deflection basins measured on the flexible pavements. Backcalculating pavement layer moduli are well-accepted procedures for the evaluation of the structural capacity of pavements. The ultimate aim of the backcalculation process from Nondestructive Testing (NDT) results is to estimate the pavement material properties. Using backcalculation analysis, in-situ material properties can be backcalculated from the measured field data through appropriate analysis techniques. In order to backcalculate reliable moduli, deflection basin must be realistically modeled. In this work, ANN was used to model the deflection basin characteristics and GA as an optimization tool. Experimental deflection data groups from NDT are used to show the capability of the ANN and GA approach in modeling the deflection bowl. This approach can be easily and realistically performed to solve the optimization problems which do not have a formulation or function about the solution.

Backcalculation of pavement layer thickness using data mining

Pavement deflection data are often used to evaluate a pavement’s structural condition nondestructively. Pavement layers are important parameters in view of bearing capacity. Pavement layer thickness may be known from the design project or site investigation. At the same time, using backcalculation analysis, flexible pavement layer thicknesses together with in situ material properties can also be backcalculated from the measured field data through appropriate analysis techniques. Data mining (DM) process has not been used as a backcalculation tool before. In this study, DM process is used in backcalculating the pavement layer thickness from deflections measured on the surface of the flexible pavements. Experimental deflection data groups from NDT are used to show the capability of the DM process in backcalculating the pavement layer thickness and compared each other. Performing the study, modeling with Kstar method gives fine results with respect to other DM modeling techniques. Backcalculation of pavement layer thickness using DM process has been carried out for the first time.

Backcalculation of Pavement Layer Thickness and Moduli Using Adaptive Neuro-fuzzy Inference System

Efficient and economical methods are important in determination of the structural properties of the existing flexible pavements. An important pavement monitoring activity performed by most highway agencies is the collection and analysis of deflection data. Pavement deflection data are often used to evaluate a pavement’s structural condition non-destructively. It is essential not only to evaluate the structural integrity of an existing pavement but also to have accurate information on pavement structural condition in order to establish a reasonable pavement rehabilitation design system. Pavement structural adequacy is often evaluated by calculating elastic modulus of each layer using the so-called “backcalculation”. Backcalculating the pavement layer properties is a well-accepted procedure for the evaluation of the structural capacity of pavements. The ultimate aim of the backcalculation process from Nondestructive Testing (NDT) results is to estimate the pavement material properties. Using backcalculation analysis, flexible pavement layer thicknesses together with in-situ material properties can be backcalculated from the measured field data through appropriate analysis techniques. In this study, adaptive neural based fuzzy inference system (ANFIS) is used in backcalculating the pavement layer thickness and moduli from deflections measured on the surface of the flexible pavements. Experimental deflection data groups from NDT are used to show the capability of the ANFIS approaches in backcalculating the pavement layer thickness and moduli, and compared each other.

Prediction of the marshall stability of reinforced asphalt concrete with steel fiber using fuzzy logic

In this study, Marshall Stability MS of steel fiber reinforced asphalt concrete has been predicted using steel fiber rate 0%, 0.25%, 0.50%, 0.75%, 1.0%, 1.5%, 2.0% and 2.5%, bitumen content 5%, 5.5% and 6.0% and unit weights 2,465--2,515 gr/cm3 by Fuzzy Logic FL. Results have shown that developed FL model has a strong potential for predicting the MS of asphalt concrete without performing any experimental studies.

Estimation of Modal Shift Potential for a New Form of Dial-A-Ride Service

The concept of a dynamic and flexible Intelligent Subscription Bus Service (I-Service) was developed, and two integrated questionnaires were conducted among the commuters of a large university camp ...

Modeling Marshall Stability of light asphalt concretes fabricated using expanded clay aggregate with Artificial Neural Networks

In this study, an Artificial Neural Network (ANN) model has been developed to estimate Marshall Stability (MS) of lightweight asphalt concrete containing expanded clay. In the model, amount of bitumen (%), transition speed of ultrasound (μs), unit weight (gr/cm3) were used as inputs and Marshall Stability (kg) was used as output. Developed ANN model results and the experimental results were compared and good relationship was found.