The United States spends $150 billion per year to maintain its vast transportation network of 4.11-million lane-miles of roads. Vehicles that drive along these roads consume 213 billion gallons of fuel per year. Due to such staggering financial and environmental costs, it is important for transportation planners to optimize current infrastructure. In California, pavement maintenance is extremely costly. Effective pavement management systems take into account cracking and roughness. The quality of these features directly impacts vehicle fuel consumption. This research has two main objectives: 1) Developing predictive performance models needed for optimized network level management of rigid transportation infrastructure. These performance models are developed by conducting big data analysis of field data collected from Caltrans Pavement Management System. 2) Calibrating MechanisticEmpirical Design Guide for rigid pavements in California and developing models for longitudinal cracking specific to California and few other states.
