Effects of initial data variability on pavement condition index and prediction of pavement deterioration

Abstract

There are several investigations that make comparisons between pavement condition assessment methods and the application of deterioration prediction models, without taking into account accurate data collection and the influence this may have on their results. It is for this reason that a research focused on evaluating the variability of results by taking data in the field was proposed. Variability is measured with the help of Pearson's coefficient of variation. To meet the objective, a field evaluation tool was developed for the purpose of collecting data to evaluate the variability of results.  In the processing and subsequent interpretation of data it was found that there is variability of results in the method, but not in the prediction model. When evaluating the results of the method it was found that the factor that influences the existence of variability is the number of faults (19 faults) and through a hierarchical analytical process (AHP) using criteria such as the frequency with which each fault is found, the degree of affectation and the percentage of each fault found, it was possible to reduce to 9 faults to be evaluated specifically in the urban area.