A new medical device can take years to develop from early concept to product launch. The long development process can be attributed to the severe consequences for the patient if the device malfunctions. As a result, three approaches are often combined to mitigate risks: failure modes and effects analysis (FMEA), simulation and modeling, and physical test programs. Although widely used, all three approaches are generally time consuming and have their shortcomings: The risk probabilities in FMEA’s are often based on educated guesses, even in later development stages as data on the distribution of performance is not available. Physical test programs are often carried out on prototype components from the same batch and, therefore, may not reveal the actual distribution of actual running performance. Finally, simulation and modeling are usually performed on nominal geometry—not accounting for variation—and only provide a safety factor against failure. Thus, the traditional use of safety factors in structural analysis versus the probabilistic approach to risk management presents an obvious misfit. Therefore, the aforementioned three approaches are not ideal for addressing the design engineer’s key question; how should the design be changed to improve robustness and failure rates. The present study builds upon the existing robust and reliability-based design optimization (R2BDO) and adjusts it to address the aforementioned key questions using finite element analysis (FEA). The two main features of the presented framework are screening feasible design concepts early in the embodiment phase and subsequently optimizing the design’s probabilistic performance (i.e., reduce failure rates), while using minimal computational resources. A case study in collaboration with a medical design and manufacturing company demonstrates the new framework. The case study includes FEA contact modeling between two plastic molded components with 12 geometrical variables and optimization based on meta-modeling. The optimization minimizes the failure rate (and improves design robustness) concerning three constraint functions (torque, strain, and contact pressure). Furthermore, the study finds that the new framework significantly improves the component’s performance function (failure rate) with limited computational resources.