Components within jet engines are subjected to extreme conditions with high temperature and mechanical load. The fatigue life predictions is an important tool and have to be the best possible. The aim of this study was to increase the knowledge about the notch effect and utilize a state of art method to predict the life of 143 small notched specimens made by a Ni-based super alloy. An initial literature survey revealed three different concepts; Highly stressed volume, Stress gradient method and Theory of critical distance. Since each concept consisted of several methods this thesis has focused on the critical distance to limit the work. Of these methods one could be adopted without further fatigue testing and development; An elasto-plastic reformulation of the Theory of critical distance. The method utilizes the elasto-plastic stress at a constant distance from the surface to predict the fatigue life of the component. Further evaluated in this thesis are; the validity of the fundamental assumptions using cyclic elastic-plastic Finite Element analysis and the predictive capability using a simplified analysis process with Neuber correction. The work concludes that, further testing is required to use the remaining methods presented in this thesis as they are, but, there might be possible to reformulate these to fit within GKN application. Regarding the adopted method the average predictions converged towards the actual life, but, further breakdown revealed that the actual life of the small notches was shorter than the predicted, which neither makes it appropriate or safe to introduce. Another finding, based on the general trend in the predictions, is that the half critical distance also reduces conservatism from the life predictions but also produce notch independence. I.e. the critical distance hypothesis shows good potential, but the critical distance needs to be determined in a different manner than investigated in this thesis work to be able to use it in conjunction with GKN design practice. This means that a constant critical distance can be used within the real projects to increase the product performance, but to further ensure durability it is essential to investigate the effect on real components.