This thesis introduces a method to determine strut-tie models in reinforced concrete structures using evolutionary structural optimization. Even though strut-tie model is considered as a basic tool in the design and detailing of reinforced concrete members subjected to bending, shear, and torsion, it is also true that these conventional methods can hardly give correct models in concrete members with complex loading and geometry conditions which have a lot of discontinuities. To overcome this problem, introduction of new approaches and modification of conventional methods are needed in determining the most efficient strut-tie models. In this thesis, the idea of evolutionary structural optimization (ESO) method is used to determine strut-tie models on the basis of the fact that the optimum topology of structures by ESO method is often a truss-like structure. Therefore, it is appropriate to apply this technique in finding the best strut-tie models in reinforced concrete structures. Several examples are provided to demonstrate the capability of the proposed method in finding the best strut-tie models. It is shown that the introduced method is supported by the existing analytical solutions and experimental evidence, and can be used effectively in practice, especially in design of complex reinforced concrete members where no previous experience is available.