| Analysis tools in pavement management systems are critical to assist transportation agencies in developing the most adequate maintenance and rehabilitation program. The significant elements are not only the type of treatment selected and its cost, but also the quantification of its short-term and long-term effectiveness and the timing of intervention. Methodologies with complex statistical approaches and performance model theories have been previously implemented in their derivation, but they may not be suitable or accurate. Therefore, it is imperative to factor in engineering experience in their assessment. In this thesis, a hybrid modelling approach was presented to systematically introduce expert knowledge in the statistical processes for quantifying the maintenance and rehabilitation short-term and long-term effectiveness, performance models, and optimal treatment times. The approach was applied to asphalt pavements using historical performance information from the Texas Department of Transportation (TxDOT) located in the Pavement Management Information System (PMIS). Furthermore, the hybrid approach was validated in the quantification of the short-term effectiveness of different construction treatments, under changing pavement characteristics, through extensive state-wide field performance monitoring. The presented hybrid model better reflected the treatment's effectiveness measures observed in the field when compared to the previous recommendations. |
