A framework for developing optimal pavement life-cycle activity profiles
Highway pavement managers seek cost-effective maintenance and rehabilitation (M&R) profiles over the pavement life-cycle. The benefits of each profile, which can be measured as the service life, increase in average performance compared to the do-nothing profile, or the area bounded by the performance curve that corresponds to the profile, may be monetized or un-monetized. The cost of each profile comprises the agency costs (of construction and maintenance) and the user costs during work zone periods and normal operations of the pavement. In a departure from existing literature on this subject, this dissertation addresses the issue of M&R optimization by developing a framework that explicitly considers the benefits and costs corresponding to each individual constituent treatment of a candidate profile. The treatment-specific performance, effectiveness, and cost models were calibrated for different functional classes and pavement types using data from in-service pavements in a Midwestern state in the United States. A nonlinear cost-effectiveness function with an integer and continuous decision variables are incorporated into mixed-integer nonlinear programming (MINLP). A number of solution techniques including outer approximation, and branch-and-bound algorithms, are proposed. The framework was applied in a case study to determine the optimal life-cycle activity profile for an interstate pavement section. Different optimal solutions were obtained for different preservation scenarios: (1) only preventive maintenance treatments, (2) only rehabilitation treatments, and (3) both preventive maintenance and rehabilitation treatments. It was found that the optimal profile, for a two-stage preservation scenario with both preventive maintenance and rehabilitation treatments, consists of thin hot-mix asphalt (HMA) overlay at the 11th year followed by a functional HMA overlay at the 20th year after the initial construction over a 32-year analysis period. It was observed that the optimal solution was generally robust for the range of realistic values of traffic and climate variables encountered in the study area. To investigate the consequences of the variability in the input parameters on the choice of an optimal profile, probability distributions of the input factors were established and Monte Carlo simulation was carried out. An innovative concept, namely, “M&R strategy contours”, was established to help asset managers to quickly identify optimal life-cycle strategies for an asset on the basis of given traffic loading and climatic conditions. ^
Samuel Labi, Purdue University.