Manufacturing yield maximization to minimize the production cost is one of the most important objectives of manufacturing. The manufacturing yield is the probability to manufacture products that satisfy multiple performance requirements. For a given set of product performance requirements, its manufacturing yield can be estimated based on the system parameter uncertainties including manufacturing tolerances, which are related to the production cost. We employ three performance indices for the design of a MEMS accelerometer, and the manufacturing yield and cost are estimated using statistical uncertainty analysis methods. The effects of the MEMS accelerometer parameter uncertainties on the manufacturing yield and cost are investigated.