Abstract As stability of perovskite solar cells remains a significant research topic, it is important to be able to predict the long-term stability of any new kinds of perovskite solar cells when new perovskite absorber materials or transport layers or new cell structures are being demonstrated. This work reports a reliable method of determining degradation rate which is resulted from thermal stress. By incorporating three kinds of accelerated tests, the activation energy for photo-thermally driven degradation processes of perovskites solar cells was determined, which is then used to predict its long-term stability using an Arrhenius equation. In addition, thermal stability of CH3NH3PbI3, HC(NH2)2PbI3, PTAA (poly[bis(4-phenyl)(2,4,6-trimethyl phenyl)amine]) and Spiro-OMeTAD (2,2',7,7'-Tetrakis[N,N-di(4-methoxyphenyl)amino]−9,9'-spirobifluorene) are studied. The thermal stability of a planar HC(NH2)2PbI3/PTAA device is better than a planar HC(NH2)2PbI3/Spiro-OMeTAD device which in turn is better than a planar CH3NH3PbI3/Spiro-OMeTAD device due to better thermal stability of HC(NH2)2PbI3 and PTAA. It is predicted that a planar HC(NH2)2PbI3/PTAA device can have a lifetime of more than 3 years (or 1.5 years) at room temperature if 50% (or 25%) drop in power output can be tolerated. While these lifetimes are specific to perovskite material chosen, preparation method and solar cell design, the lifetime prediction method reported here can be verified experimentally. Therefore, the lifetime calculation method developed in this work is a quick and useful tool for determining the relative stability of a perovskite device especially when comparing the merits of different cell structure designs.