During accelerated degradation testing (ADT), the component is exposed to higher levels of parameters that cause it to fail faster than it would under normal operating conditions. This approach saves time and reduces expenses associated with tool life tests for valuable workpieces. The fundamental assumption of ADT is that failure modes observed at elevated levels of parameters are similar to those under recommended usage conditions. In this study, tool failure modes are utilized to determine the appropriate levels of predictor variables, like cutting parameters and tool nose radius, for ADT of PVD-coated carbide tools during dry machining of Inconel 800. To study tool failure modes, experiments are done at three levels of cutting speed (V_c), feed rate (f), depth of cut (a_p), and nose radius (r) by following the Taguchi L₉ orthogonal array. Notably, a favourable tool failure pattern is observed when employing the middle level of the nose radius and operating at relatively higher cutting speeds while ensuring that f and a_p fall within the recommended range. In the future, with the parameters’ levels established through this study, full-scale ADT will be performed and extrapolated to predict the reliability function of the cutting tool under normal conditions. The suggested method holds potential for various applications, including estimating the reliability of cutting tools and monitoring tool condition, among others.