A probabilistic approach to the life-cycle cost assessment of preventive maintenance strategies on flexible pavements.

摘要

Preventive maintenance has been recognized as a tool that has the potential to reduce expenditures for the operation of pavement networks. Properly specified pavement deterioration performance models are essential for accurate predictions of future pavement conditions, which are crucial for identifying the least-cost maintenance and rehabilitation strategies that maintain the desired levels of pavement condition. Pavement deterioration is a complex mechanism which involves not only structural damage but also the interaction between traffic, material, environment and time. Pavement crack initiation models are among the most complex of these performance models.;Past research investigations have based their cost analyses on deterministic pavement deterioration prediction models, which lack the ability to capture the random part of the pavement's true field behavior. Empirical-mechanistic models based on observations of in-service facilities allow a more realistic representation of pavement performance. Developing such models, however, presents a more challenging problem due to the many limitations of field data, including measurement errors, data censoring, and missing observations.;In this dissertation, empirical-mechanistic pavement crack initiation models are developed based on stochastic hazard rate modeling techniques. Pavement condition survey data for the highway system in the State of California is used for the model development to allow representation of the effects of a wide range of facilities and environmental conditions.;Once the models were developed, several scenarios for pavement maintenance were compared with the control case of rehabilitation only. These performance results were used with available cost information to develop life-cycle cost comparisons.;This research confirms the notion that pavement preventive maintenance practices do reduce long term roadway maintenance costs.