Thermal barrier coatings (TBCs) are widely used to protect gas turbine blades but internal stress near the interface in TBCs is one of the main causes of thermal barrier failure under thermal cycling. A non-destructive inspection technique based on Eu3+ photoluminescence piezospectroscopy has been successfully used to analyze the residual stress in TBCs, but systematic and quantitative evaluation of data processing is still needed, especially with respect to the identification of peak positions. In this work, processing methods for Eu3+ photoluminescence spectroscopy data were studied to characterize TBC internal stress. Both physical and numerical experiments were carried out where Eu3+ luminescence spectra were obtained from a sample of europium-doped yttria-stabilized zirconia (YSZ:Eu3+) under step-by-step uniaxial loading, and the simulated spectra were numerically deduced from the measured spectra. The peak shifts were then obtained by processing the spectral data in different ways (Gaussian, Lorentzian, pseudo-Voigt fitting, and the barycenter method), and comparing the results. We found that the Gaussian function, rather than the commonly used Lorentzian function, is the most appropriate method for the application of Eu3+ photoluminescence piezospectroscopy in TBCs because it provides sufficient sensitivity, stability and confidence for quantitative stress analysis.