Automotive wheel bearings have the primary function of translating the rotating motion of the wheels into linear vehicle motion while supporting the vehicle weight. As vehicle lives continue to increase, there is a need for longer service lives than those of existing products. There is an even greater need for performance-related reliability during usage. Lateral stiffness, one of the main parameters of wheel bearing design, has a significant influence on ride comfort and steering feel. In this study, reliability-based weight optimization considering geometric uncertainty for automotive wheel bearings was investigated. Deterministic design optimization (DDO) and reliability-based design optimization (RBDO) were performed. For optimization, the following three key relationships were chosen: wheel bearing specification and geometry for design variables, weight for cost function, and stiffness for constraint. A Monte Carlo simulation considering the probability distribution of the geometric uncertainties was performed to identify the effect of variations in dimensions. Since Monte Carlo simulation to analyze the stiffness of the bearings takes a very high computational cost, a regression analysis was performed, followed by running Monte Carlo simulation based on this regression. Based on the simulation results, the design variations for the automotive wheel bearings were evaluated using the results of the Monte Carlo simulation. The results showed that the DDO and RBDO reduced weight by about 36% and 17% respectively, compared with their initial weight.