Evaluation of the Pavement Condition Index Prediction Model for Flexible Airfield Pavements.

摘要

The U.S. Army Construction Engineering Research Laboratory (CERL) is in the process of developing a comprehensive pavements maintenance management system for the Air Force. A major step in the development was the development of models capable of predicting the future Pavement Condition Index (PCI) of a pavement feature, based on environmental and situational variables. These models are still considered tentative. The purpose of this thesis was to evaluate the current model proposed for predicting the PCI in flexible airfield pavements. The current model was found to be based on a statistically questionable change in the data. The current model was reaccomplished using a more accepted practice. The reaccomplished model was evaluated against a new data base and found to reasonably predict the PCI. The new and existing data was then combined, the model again was reaccomplished, and improvements noted. Two models were developed that show the possibility of interaction among the variables and the presence of nonlinear relationships. Finally, a model was developed, based on only the new data, which predicts the PCI very accurately for pavement on the base from which the data was gathered. This research should supplement CERL efforts to validate their models. (Author)