Resumen
This paper introduces the concept of targeted field investigation on the reliability of earth-retaining structures in an active state, which is implemented in a random finite element method (RFEM) framework. The open source RFEM software REARTH2D was used and modified suitably in order to accommodate the purposes of the present research. Soil properties are modeled as random fields, and measurements are modeled by sampling from different points of the field domain. Failure is considered to have occurred when the ?actual? resultant earth pressure force on the retaining wall (calculated using the friction angle random field) is greater than the respective ?predicted? force (calculated using an homogenous friction angle field characterized by the mean of the values sampled from the respective random field). Two sampling strategies are investigated, namely, sampling from a single point and sampling from a domain, through an extensive parametric analysis. As shown, the statistical uncertainty related to soil properties may be significant and can only be minimized by performing targeted field investigation. Among the main findings is that the optimal sampling location in the active state is immediately adjacent to the wall face. In addition, it is advisable that the entire wall height be considered in sampling. Finally, it was observed that the benefit from a targeted field investigation is much greater as compared to the benefit gained using characteristic values in a Load and Resistance Factor Design framework.