非破壞試驗(NDT)儀器被廣泛地用來評估既有鋪面之現況，鋪面各層的彈性模數值通常係由面層撓度資料回算而得。因鋪面撓度量測值會因許多因素不同而異，本研究主要目的在探討有限尺寸與荷重傳遞效應對剛性鋪面回算之影響。 本研究首先將對AASHTO所建議的二層剛性鋪面回算流程進行分析驗證。雖然該流程中已考量有限尺寸的修正方式，然而其對於現地鋪面回算之適用性問題則有待進一步的討論。本研究係利用ISLAB2000有限元素程式來協助後續的分析，並透過執行一系列的有限元素程式來建立鋪面的結構反應資料庫。 理論分析結果發現有限尺寸與荷重傳遞效應對於所產生的撓度值有明顯的影響。版尺寸較大時，最大撓度值會接近Westergaard的無限版理論解。當版的尺寸變小時，AASHTO所建議的有限尺寸修正方式確有其必要性，否則回算結果將可能被低估。而荷重傳遞效應會使短版塊間產生連鎖的效應，回算分析時可將其視為無限版塊，毋須做有限尺寸修正。 再者，本研究並擷取美國長程鋪面績效(LTPP)資料庫之現地回算資料，再與實驗室的彈性模數值相互比對。結果發現其間仍存在有相當高的變異性，亦指出仍需要對現有的最新回算技術做更深入的研究。研究中亦發現對於現地鋪面之回算問題，不論是長版塊或短版塊，或許是因為版塊間有荷重傳遞效應的存在，並無有限尺寸修正的必要。 Non-Destructive Testing (NDT) devices have been widely adopted to evaluate existing pavement conditions. The surface deflection data was often used to backcalculate the elastic moduli of pavement layers. Since there existed many factors affecting the deflection measurements, the primary objective of this study was to investigate the effects of finite slab sizes and load transfer efficiency on rigid pavement backcalculation. The two-layer backcalculation approach proposed by the AASHTO for the structural evaluation of existing rigid pavements was first investigated. Even though the proposed approach has already provided adjustment factors to account for the effects of finite slab sizes, their applicability to in-service pavement backcalculation problems was yet to be further discussed. The ISLAB2000 finite element (FE) program was adopted to assist in the following analyses. Structure response databases were created through a series of FE factorial runs. Theoretical investigation indicated that finite slab sizes and load transfer efficiency did have significant effects on the resulting FE deflections. For longer slab, the maximum deflection is close to the Westergaard infinite slab solution. For shorter slab, the AASHTO’s proposed adjustment for the finite slab size effects is indeed necessary; otherwise the backcalculated layer moduli might be significantly underestimated. Since load transfer efficiency will result in interlocking effects on shorter slabs which can be treated as infinite slabs, no slab size adjustment on backcalculation results is needed. Furthermore, the laboratory tested layer moduli were compared with the backcalculated moduli using the Long-Term Pavement Performance (LTPP) database. Relatively high variability between them was observed indicating that further research study is needed to improve the current state-of-the-art backcalculation approach. In addition, it was also found that for practical backcalculation problems of in-service pavements, the slab size adjustment was not necessary for both shorter slabs and longer slabs, perhaps due to the fact that these pavements were interlocked by load transfer effects.