The booming market of sustainable energy increases the usage of offshore wind turbines. Simultaneously, with the rising demand for the foundation of major bridges and elevated light rail supports onshore, the total use of large diameter monopiles is roaring and expected to grow faster for the world to continue combating climate change and urbanization. The objective of this study is to address the applicability of using P-y curves for the design of large diameter pile foundation in both clay and sand. P-y curve approaches are the most common method to predict the behavior when pile foundations subjected to lateral loading. P-y curves were first suggested by McClelland and Focht (1956). Then being developed and calibrated by Matlock (1970) and Reese et al. (1974) by using comparative small diameter piles (0.3 m to 0.6 m). This study confirms the effect of pile diameter with a total of 110 lateral pile load test collaborated in both sand and clay from nine counties around the world. The comparisons were simulated in a differential equation-based software, LPILE (Ensoft Inc.). The modification to sand (API RA, 2011) is proposed by fitting the subgrade modulus k in sand when the predicted deflection at the pile top has matched or around the measurement at 33% ultimate load (H/Hou = 33%). A low R2 was obtained to show that there was no relation between the ratio of predicted and measured deflection and the pile diameter when proposed nk was employed and used in API sand approach. Besides, the modification to both soft (Matlock, 1970) and stiff clay (Reese et al., 1975) is proposed by back calculated the y50, the deflection at one-half the ultimate soil resistance with a total of 34 measurements. This modification has been validated with a three-dimensional finite element (FE) software. And a low R2 was obtained as well to confirm the elimination of pile diameter effect in clay. Moreover, the base translation spring and the base moment spring are added in an effort to better represent the behavior of the pile subjected to lateral loading. This is particularly important in the case where the pile has a small length L to diameter B ratio (L/B <= 7.0). A good agreement was found that both pile tip shear and moment resistance and pile side friction resistance contribute to the resisting system of a rigid monopile. Lastly, the collected databases are used to quantify the statistical parameters associated with the scatter between the predicted and measured pile top deflection as well as the probability that a deterministically predicted deflection will be underestimated. These results provide probabilistic factors to engineers when designing the pile foundation and ensure the reliability of the design. In sum, five objects include data collection, pile size effect detection, proposed and validated modification, extra resistances consideration, and statistical approaches were completed, and solid conclusions were addressed.
