Abstarct: Employing the computer aided design technique in the newly developed isogeometric ‎analysis method has many advantages, e.g. removing the error of geometrical modeling. On the other hand, in all numerical methods, errors due to approximation of unknown ‎functions are inevitable‎ and researchers have been concerned about the reliability of the results. This thesis is to develop methods that can be used to estimate error rate of the isogeometric analysis method. These concepts are arranged in two main parts. In the first section, the error estimation method baxsed on the superconvergent points property for isogeometric analysis of stress-strain, axisymmetric and three-dimensional problems have been considered; and will show why the Gauss integral points has convergence properties in the isogeometric analysis method. In this method, by making use of the superconvergent points, each of the components of the improved stress tensor is considered as an imaginary surface. This surface is generated by using the same NURBS’ basis functions that are employed for approximation of the primary variable in the isogeometrical analysis. The obtained results of all examples indicate the effectiveness of the proposed method in the estimating the error of isogeometric analysis of the stress-strain, axisymmetric and three-dimensional problems. In the second part, a new method is introduced for calculation of stress field in the isogeometric analysis method that makes use of equilibrium of patches. In this technique, by considering the forces induced on the patches of isogeometric analysis, the surface denoting the variations of each component of the stress tensor is approximated by the same order of NURBS’ shape functions that are used for approximation of the displacements. One of the useful features of this method is being independent of the Gauss integration points, that is especially advantageous when a different integration method than the Gauss quadrature is employed. To demonstrate the performance of the method, six examples are taken into consideration and their exact and approximate error energy norms are calculated. The obtained results indicate that in all of the considered examples error estimation by this approach is superior to our previous method baxsed on using the superconvergent points and therefore can be considered as a another simple and efficient method for error estimation and stress recovery in isogeometric analysis.