An automotive engine can be more efficient if thermoelectric generators (TEG) are used to convert a portion of the exhaust gas enthalpy into electricity. Due to the relatively low cost of the incoming thermal energy, the efficiency of the TEG is not an overriding consideration. Instead, the maximum power output (MPO) is the first priority. The MPO of the TEG is closely related to not only the thermoelectric materials properties, but also the operating conditions. This study shows the development of a numerical TEG model integrated with a plate-fin heat exchanger, which is designed for automotive waste heat recovery (WHR) in the exhaust gas recirculation (EGR) path in a diesel engine. This model takes into account the following factors: the exhaust gas properties’ variation along the flow direction, temperature influence on the thermoelectric materials, thermal contact effect, and heat transfer leakage effect. Its accuracy has been checked using engine test data. Based on the simulation results, the equation for MPO was fitted to thermoelectric material properties, which include Seebeck coefficient αTE, thermal conductivity κTE, electrical resistivity ρTE, and operating conditions, which include the exhaust gas flow rate fh and gas-in temperature TIn. It is found that the regression equation provides an excellent prediction of the MPO when the TEG is used in EGR path. The sensitivity analysis based on the regression equation revealed that MPO is most sensitive to αTE and TIn, and least sensitive to fh. It has also been discovered that the MPO is relatively less sensitive to the fh when TEG is working at high flow rates.