| This study resulted in a comprehensive remaining life model for Continuously Reinforced Concrete Pavements (CRCPs), which simulates the effect of traffic and environmental damage on CRCP, and predicts the time when the pavement will structurally fail. Theoretical and experimental knowledge were used to develop the model and a database with CRCP performance data, which was used to calibrate the model, then the ealibrated model was applied to other CRCP sections. The remaining life model was evolved from several steps: (a) concept of the remaining life; (b) mathematical modeling of the concept; (c) sensitivity analysis of the model; and (d) computer modeling and calibration of the model with actual data. The concept of the remaining life is focused on the response of CRCP to traffic and environmental loads, and the relationship of the external loads with the remaining life and on the phases of this response--early age, stable period, and failure period. Mathematical modeling included a detailed application of partial derivative method for standard deviation calculation, which was used in reliability assessment of remaining life model. Then, a sensitivity analysis of the remaining life model was applied to check the correctness and accuracy of the equations and procedures. The computer implementation of the model, Remaining Life Model for CRCPs (REMLIF) was created and calibrated with CRCP performance data from the Center for Transportation Research database. The calibration resulted in a set of coefficients for annual minimum temperature, annual total rainfall, and coarse aggregate type. The calibrated model was applied to test sections located in Houston, Texas. The model application is followed by a chapter on how to implement and interpret the remaining life model and its output, respectively. The remaining life model offers several improvements over currently available CRCP performance models: (a) contribution of stabilized subbases and environment to CRCP performance are modeled; and (b) reliability assessment is done with the contribution of all of the significant variables. As a result, three different output data are provided for (a) maintenance scheduling; (b) overlay design; and (c) punchout prediction of the analyzed CRCP. |
