Standard Engineering Principles in PMS Applications

Abstract

An essential element of any Pavement Management System (PMS) is its ability to correctly predict the future “condition” of pavements. When a new PMS is introduced to a highway agency, the likelihood of successful implementation and acceptance is strongly dependent upon how well future pavement condition, as predicted by the system, agrees with local engineering knowledge of those pavements. Highway engineers still play an important role in the implementation of a PMS. Modeling pavement performance is extremely complicated, and no PMS can consider more than a few of the parameters involved, and then only in a highly simplified manner, whether they be analytical or “expert” system techniques. In order to improve the objective engineering performance and avoid rejection of the system, it is suggested that the following capabilities should be included in any PMS: ● A PMS must be capable of predicting structural as well as functional deterioration. Materials in the pavement layers, and their degradation under the effects of time and loading, are of primary concern to those highway engineers responsible for the maintenance and performance of the pavement network. A system that empirically predicts “only” the future ride quality or user costs ignores one of the main engineering considerations, and may be rejected as being too subjective or “political”. Currently accepted performance prediction techniques are usually applied at the project level for design purposes. At the network level, their application becomes somewhat more difficult, but now these difficulties can be dealt with. ● A method must be provided which uses predictive models tied to historical data as it becomes available in the specific PMS database, and for modifying these models so that the engineers' knowledge of local materials, environmental effects, construction and maintenance practices, etc., can easily be incorporated in the modeling procedure. In other words, predictive models based on sound engineering principles can be calibrated for a specific PMS application using historical data from the system itself. “Modeling” pavement performance by extrapolating future condition from historical data is a technically unacceptable simplification, however, because the effects of material degradation, maintenance, or rehabilitation measures cannot be considered. The engineering techniques for dealing with this new approach already exist and are applicable to a system-wide analysis. Statistical use of historical data (often of dubious quality) appears to be a poor substitute for engineering skill and the use of presently available analytical tools.