鋪面分析中由於材料本身的變異性及施工的不確定性，使其分析結果亦為隨機性，使用隨機性的分析方法將可對分析結果提出更可靠的依據。本研究之主要目的是發展隨機有限元素法（stochastic finite element method, SFEM）於鋪面之分析，並將隨機有限元素法所求得之結果與蒙地卡羅模擬法（Monte Carlosimulation technique）加以檢驗所發展程式之正確性，並探討鋪面材料變異性對鋪面反應統計參數的影響。本研究亦使用可靠度理論建立功能函數，並與隨機有限元素之結果結合發展具有可靠度之鋪面績效模式。研究中建立三種不同之鋪面結構與材料統計參數，並利用結構分析所得反應的統計參數與可靠度理論結合，探討使用一階近似法與一階可靠度法（first order reliability method,FORM）對鋪面績效模式可靠度因子的差異，此績效模式之可靠度因子將可作為鋪面績效的可靠度分析。

Due to the variations of material properties and construction in pavement engineering, pavement responses are stochastic. The use of stochastic analysis in pavement analysis is reasonable. The purpose of this study is to develop a perturbation stochastic finite element program in pavement analysis and our results are compared with the results of a Monte
Carlo finite element program. The results of the stochastic finite element
analysis are incorporated with the reliability method to develop performance function models. The reliability-based performance models will be used for performance analysis of test pavement sections. The first order approximation method and first order reliability method (FORM) are used to compare the factor of reliability for rutting and fatigue cracking of pavement. The results of these analyses can be used to study pavement performance reliability.