For several years, the MATIERE company has developed a new type of reinforcement based on the use of ribbed flat steel in substitution of high-adhesion round steel used for precast products. This work is a part of the Technological Research and Development program implemented in the frame of Marcel MATIERE patents. The interest of these new ribbed carbon flat steel lies in their geometry that allows to consider a new structural disposition associated with a gain on the thickness of concrete, particularly at the coating level. These steels are mainly used for thin elements where they will achieve the savings of the most significant concrete. However, no standard or regulation take into account in this time for these new steels. A scientific study validated, mainly based on numerical modeling, is so necessary, firstly, to model and understand the interaction between the flat steel and the concrete, and secondly, to justify the traditional calculation methods are applicable to such reinforcement. In this frame, a modeling strategy based on a probabilistic multiscale approach was developed. This multiscale approach is not to develop a model that includes, in its formalism, all modeling scales, from the very local to the global scale (complete structure), but to develop a panoply of models that provides the relevant informations on the scale of analysis chosen. Whatever the scale considered, the modeling can provide the information relative to the cracks opening and cracks spacing. The probabilistic aspect is mainly due to the heterogeneity of the concrete. The developed models can take into account the scale effects, specific heterogeneous materials, which play an important role in the behavior of concrete structures. The research was therefore to develop a modeling tools of the interface behaviour which is perfectly consistent with the scale modeling of phenomena considered, particularly with regard to the cracking process of structure reinforced by steel plates. The scientific approach was based on a parameters identification of the modeling by an inverse analysis performed on the results of experimental tests carried out on the tie-beam reinforced by flat steel. Modeling tools were then validated on the modeling of the beams-slabs in flection and compared to experimental results. All experimental tests for this study were made by Polytech Clermont Ferrand to the request of the MATIERE company
