In the past few decades, there have been an improved awareness of environment sustainability to bring down the effects of climate change. The conventional energy generation via fossil fuels are being shifted to renewable energy sources. Alongside with this global trend, the automotive industry is undergoing a rapid transformation towards electrical drives (E-mobility). Such high power electrical drives consists of numerous electrical components that operate under severe thermomechanical loads. The high integration densities of components in the electrical systems (such as converters, inverters, etc) result in high heat dissipation per unit area and its intercoupling to adjacent components failures. A comprehensive analysis of the crucial electrical components is essential to ensure its reliable operation. Owing to the present extreme competitive time to market requirement of automobiles, numerical simulations are highly useful to extensively study the impact of various loads, physical and geometrical properties. Thereby, some experimental procedures can be avoided, which reduces cost of testing and improves time to market. The failures occurring in the electrical systems are thermo-mechanical in nature. The most common failures observed in the power systems are die-attach failure, bond wire lift-off, and metallization crack on the die and mold compound crack. Therefore, lifetime estimation of the electrical system is necessary to evaluate the performance of the system. This is performed through numerical simulations and experiments. By this estimation, an analytical model is developed to predict the systems lifetime before the products development. Initial lifetime prediction will be based on experiment and simulation as both results should have a good agreement with each other. This paper focuses on analyzing the reliability issues in power electronics by finite element analysis (FEA) at the very beginning of the development process. This approach's major goal is identifying system behavior trends to support design decisions and enable early concept evaluations. To take into account existing uncertainties regarding material and geometric structure of the device a DoE study was conducted to perform a sensitivity analysis. The lifetime estimation of the system was based on the Coffin-Manson equation which consists of accumulated cyclic equivalent plastic strain, extracted from the numerical simulations, and the number of mean cycles to failure, obtained by experiments of comparable systems. The accumulation of plastic strain in the system was caused by stresses generated in the system during the thermal loading. In this investigation material properties, such as yield strength, Young’s modulus, coefficient of thermal expansion, etc. and geometric properties were varied to investigate the most influential parameters for the strain accumulation in the system. This helps the process and design engineers to select appropriate material and geometry configurations for developing more reliable systems according to the given requirements.