Eddie Yein Juin Chou

Automated pavement distress detection using advanced image processing techniques

In this paper, a novel, fast and self-adaptive image processing method is proposed for the extraction and connection of break points of cracks in pavement images. The algorithm first finds the initial point of a crack and then determines the cracks classification into transverse, longitudinal and alligator types. Different search algorithms are used for different types of cracks. Then the algorithm traces along the crack pixels to find the break point and then connect the identified crack point to the nearest break point in the particular search area. The nearest point then becomes the new initial point and the algorithm continues the process until reaching the end of the crack. The experimental results show that this connection algorithm is very effective in maximizing the accuracy of crack identification.

Multiyear pavement repair scheduling optimization by preconstrained genetic algorithm

Development of a genetic algorithm (GA)–based optimization tool for determining the optimal multiyear pavement repair schedule is described. The problem of finding the best multiyear work plan can be modeled as a combinatorial optimization problem, the objective of which is to achieve the highest possible average network condition for a given budget and operating constraints. Two GAs and a dynamic programming (DP) approach were implemented to determine multiyear repair schedules. The DP method resulted in optimal solutions, but it suffers from rigidity and inability to handle large-scale problems. The GA techniques obtained solutions that were near optimal and maintained flexibility and scalability. The GAs implemented were of two types: simple and preconstrained. The simple GA uses constraints when searching for solutions. The preconstrained GA uses constraints that limit which repair can be selected before searching for solutions to attain better efficiency.

Development of speed-related ride quality thresholds using international roughness index

The International Roughness Index (IRI) has been used as a road roughness index on highways for nearly two decades. Research shows that subjective ratings such as present serviceability rating or mean panel rating could be reasonably predicted from IRI. Acceptable IRI values for highways have been developed and published by FHWA. Such IRI standards, however, are not applicable to local streets. Because vehicle speeds are lower on local streets, a comfortable ride can still be achieved at higher IRI values. There is a need to determine the acceptable IRI threshold values for local streets with various speed limits. Such IRI threshold values allow local pavement management officials to compare objectively the ride quality of streets with different speed limits. It was found that human ratings depend linearly on the logarithm of the rate of change of the vertical acceleration, namely, the jolt or jerk experienced by the raters. By analyzing the IRI records in the Long-Term Pavement Performance database and the corresponding calculated jolts, the authors found that jolt is linearly proportional to IRI at a given speed and approximately linearly proportional to the travel speed for a given IRI. By assuming further that the same jolt will lead to the same ride quality and that Interstate highways are operated at 120 km/h (75 mph), the authors use the jolt corresponding to the IRI thresholds set by FHWA for highways to develop speed-related ride quality thresholds at different travel speeds. The variation of jolt for roads with the same IRI value is also discussed.

Pavement Condition Prediction Using Clusterwise Regression

In this paper, the clusterwise regression model, which fits data to more than one curve, was introduced to model the deterioration of pavement condition. To make the model solvable in practice, a modification was made to the model by the membership of a pavement to each cluster being estimated with the fuzzy sets concept with the corresponding errors. The number of unknowns in the modified model is reduced significantly. The model was then extended to be applicable to cases in which more than two clusters or nonlinear equations were used. Based on the result of the modified clusterwise regression model, a procedure was proposed to predict the pavement condition rating (PCR) for any individual pavement, given that the pavement condition rating at the present age was known. In the example, the ordinary least squares regression method was first employed to determine the PCR prediction curve for a pavement group. The PCRs of the individual pavements were predicted by an adjusted prediction curve based on the prediction curve of the group. The proposed prediction procedure was then applied to the same data set to make the predictions. The results showed that the proposed procedures using the modified clusterwise regression method could result in a smaller prediction error and thus produce a more accurate prediction than one produced with the adjusted curve. Therefore, to improve the accuracy of the pavement condition predictions, the modified clusterwise regression model is recommended in cases in which the pavement families are not well defined.

Pavement information system: Detection, classification and evaluation

This paper presents a method for pavement crack detection, classification and evaluation using the Radon transform. The detection part of the algorithm is built upon the wavelet transform and the evaluation part is considered in the Radon transform domain. Since cracks have specific linear features in the space domain, the Radon transform can effectively be used on a binary image to classify and evaluate the cracks based on their possible patterns. The proposed algorithm assumes the fact that crack pixels are darker than their surroundings. The simulation results show that the proposed method is very effective and can provide information about the type and severity of the pavement cracks.

Cross-Asset Transportation Project Coordination with Integer Programming and Constraint Programming

Many asset management systems generate work plans for different asset components independently. Coordination of projects between different assets, such as pavement and bridge rehabilitations in the same, adjacent, or nearby locations, is often performed only on an ad hoc basis. Poorly coordinated projects might cause unnecessary traffic disruptions and increased agency costs. This paper presents two optimization models to coordinate asset projects systematically with the application of integer programming and constraint programming. The objective of the two models was to maximize the total benefits of the projects, including project benefits and coordination benefits, by rescheduling some projects subject to budget constraints and project timing constraints. The project benefits were assessed on the basis of relationship between the asset condition and the vehicle operating cost. The project coordination benefits were estimated for road users and highway agencies. An example problem was solved to improve an uncoordinated work plan consisting of 26 asset preservation projects. The results show that the two models yield the same solution and the total benefits could be increased significantly by project coordination. The models can be implemented by highway agencies to support their cross-asset management decisions.

Calibration of Fatigue Cracking Model of Flexible Pavements Using Mechanistic-Empirical Based Probabilistic Method

In order to utilize a mechanistic-empirical procedure in flexible pavement design, a procedure being able to predict the occurrence of the fatigue cracking of pavement in the field is needed. A mechanistic-empirical based probabilistic procedure was proposed for fatigue cracking modeling in this paper. The fatigue life of the pavement was assumed to be dependent upon the means of tensile strain at the bottom of asphalt layer, resilient modulus of asphalt layer, and applied traffic loading, as well as their variances. The damage ratio D was assumed to be a log-normal distribution and the percentage of area cracked was computed as the probability that log(D) exceeded zero. Taylors expansion was employed to calculate the mean and variance of log(D) . The regression coefficients in fatigue cracking model were then determined by minimizing an objective function that represented the differences between the measured and predicted fatigue cracking areas. A procedure for converting categorical fatigue cracking ratings to numerical values was also proposed. The root mean squared error (RMSE) of the calibrated fatigue cracking model is about 7% (of the wheelpath area). The proposed procedure for the calibration of fatigue cracking model requires field and laboratory tests to verify its validity.

Estimation of the Effects of Influential Factors on Pavement Service Life with Cox Proportional Hazards Method

Estimating the effects of influential factors on pavement service life is an important, but difficult task for highway agencies. In this study, by specifying pavement condition rating (PCR) of 70 as the terminal pavement status, survival curves were developed based on historical PCR data using Cox Proportional Hazards method. Further, the estimated service lives are obtained from these survival curves. As an example, the survival data of asphalt overlays on flexible pavements in Ohio were analyzed for this study. The effects of influential factors such as structure thickness, climate, traffic loading, and pavement conditions prior to repair on pavement service life were addressed. The results showed that the Cox Proportional Hazards model is applicable in estimating the effects of influential factors on pavement service life. The service life obtained from this study can be used to assist with pavement rehabilitation decision-making, overlay design, and budget allocation.