4 th Brazilian Conference on Composite Materials. Rio de Janeiro, July 22 nd -25 th , 2018 1 ASSESSMENT OF THE LOAD CAPACITY OF FIBER REINFORCED CONCRETE Marcello Congro (1) (2) , Cristian Mejía (2) and Deane Roehl (1) (2) (1) Department of Civil and Environmental Engineering, Pontifical Catholic University of Rio de Janeiro, Brazil. (2) Tecgraf Institute, Pontifical Catholic University of Rio de Janeiro, Brazil. https://doi.org/10.21452/bccm4.2018.02.05 Abstract Prediction and verification of material properties are essential to ensure the performance of most engineering structures. For those involving composite materials, as fiber reinforced concrete, the main target of probabilistic studies is not only related to the concepts of lifespan and durability, but also to how fiber distribution affects the macro material behavior. In practice, during the concrete mixing process, steel fibers are randomly distributed and dispersed under the cementitious matrix. According to fiber arrangement, orientation, and geometry, fractures can propagate along different paths. Computational simulations are employed to predict load capacity of a given structural system. This paper presents the numerical modelling of direct tensile test for a fiber reinforced concrete. Cohesive interface elements are used to model the steel fiber behavior within the concrete matrix. These cohesive elements are placed at the edges of the solid elements, allowing fracture propagation. In order to reproduce the effect of the random distribution of steel fibers within the cementitious matrix, random values of elasticity modulus E and tensile strength are assigned to each solid and cohesive element, respectively. Marangon (2011) provides the mean and standard deviation values of the experimental data. Normal, lognormal and logistic distributions are considered for each parameter. Three distinct simulation sets are analyzed: (i) structured mesh with random elasticity modulus for each solid element, (ii) structured mesh with random tensile strength for each cohesive element, and (iii) unstructured mesh with random tensile strength for each cohesive element. For each set, the predicted fracture paths and load capacity present satisfactory results when compared to those obtained experimentally by Marangon (2011). All three-distribution functions lead to results in the expected experimental range. The results also show that the fiber dispersion and orientation contribute to the structural load capacity, increasing the structure durability. 1. INTRODUCTION