Following the basis of the ASME codes, the major nuclear components are designed to successfully avoid the fatigue failure. However, such design is generally very conservative and it is necessary to accurately assess the fatigue life of the components for the optimal life. The assessment of fatigue damage accumulation due to the thermal transients is currently performed via online fatigue monitoring systems. The algorithms for online calculation of thermal stress are one of the main components of these systems and are often based on the Green function technique (GFT), in which machine parameters such as fluid temperatures, pressures, and flow rates are converted into metal temperature transients and thermal stresses. However, since the GFT is based upon the linear superposition principle, it cannot be directly used when the temperature-dependent material properties are considered. This paper presents a methodology to consider the temperature- dependent material properties using artificial parameter method. Two cases are presented to compare the results calculated from the proposed models with those calculated by finite element method (FEM). It is found that the temperature-dependent material properties have significant influence on the maximum peak stresses which can be accurately captured by the models proposed in this work.