For the formulation of the pavement distress index based on deduct-values, the essential task is the determination of deduct-values for user-defined type-severity-density distresses so that these determined deduct-values would replicate the pavement condition ratings given by pavement experts. Traditionally, the determination of these deduct-values is an iterative trial-and-error process. In this paper, a new approach that is capable of automating this process is presented. In the proposed algorithms, the total deduct-value for a specific sample is a weighted summation of each individual deduct-value that is present. The weight associated with a specific deduct-value is calculated using a polynomial weight function of the deduct-value's component (percentage) of the total deduct-value for the particular sample. The deduct-values and the coefficients of the polynomial weight function are determined using non-linear programming techniques. The algorithms were verified and validated using simulated data from PAVER, a well-known pavement management system from the US Army Corps of Engineers. Field distress survey data from Minnesota Department of Transportation was used to demonstrate the implementation process of this approach. (a).
Abstract