Locating schemes, used to position parts during manufacturing, are usually designed in such a way that the response from the system is minimized. This implies that the position of the fasteners and/or welds are known in an assembly. Today there exist numerous of methods aiming to find an optimal set of locating points to increase the stability of an assembly, for both rigid and compliant parts. However, various industrial applications use surface-to-surface contacts to constrain certain degrees of freedom. This can lead to designs sensitive to geometric and load variations. As the complexity of the surfaces increases, difficulties of allocating geometric tolerances arise. An approach to control this is to keep the contact locations statistically stable. In this paper a methodology is presented where the First-Order reliability Method (FORM) is applied for numerical data, retrieved through Finite Element Analysis (FEA), to ensure that statistically stable contact location are achieved for two bodies with surface-to-suface contact. The FEA data represents how much of the total stress that lies within a given area, sW. The data is continuous and therefore it is assumed that the gradient can be calculated numerically with small steps. The objective function is to maximize sW for n variables. The data set is simulated through Finite Element Analysis using the commercial software Ansys and the results is illustrated on a case study from the machining industry.