Simultaneous analysis of structural reliability and parameter sensitivity of flexible pavements

Abstract

To account for the influence of uncertainties in pavement design, adopting a probabilistic limit state-based approach for asphalt pavement structural design is currently the prevailing solution. However, due to the numerous parameters involved in asphalt pavement design, the accuracy and efficiency of reliability and parameter sensitivity analysis still need improvement. This study, based on the Direct Probability Integration Method (DPIM), has developed a new synchronous analysis method for the reliability and parameter sensitivity of flexible pavement. Firstly, the principles of DPIM for calculating the reliability of flexible pavement structures are introduced. On this basis, formulas for calculating the reliability sensitivity of random parameters are derived from the perspectives of mean (μ) and coefficient of variation (CV). The numerical solution steps are also explained. Using a 6-layer flexible pavement structure, coupled with the pavement structure mechanical response analysis program (Apbi) and a calculation program developed in MATLAB, the structural reliability is calculated. The efficiency and accuracy of reliability calculations using the DPIM are compared with the Monte Carlo Simulation method (MCS). Simultaneously, the reliability sensitivity of 15 random parameters, including traffic parameters, structural layer geometry, and material parameters, is quantified. The results indicate that the DPIM can efficiently obtain the reliability of flexible pavement that meets accuracy requirements and identify the design parameters that have a significant impact on the reliability of flexible pavement. This method can improve the efficiency and quality of asphalt pavement design.

Keywords

• Reliability analysis
• Reliability sensitivity
• Direct probability integral method
• Flexible pavements