Statistical process adjustment with upside-down normal loss function

The upside-down normal loss function can accurately reflect and quantify the quality loss associated with a process in a way which minimizes the expected loss of the upside-down normal. This paper proposes a statistical process adjustment scheme for the process with upside-down normal loss function. We estimate the process output quality in a Bayesian framework, and then derive the proposed adjustment scheme in detail. An example is given to illustrate the performance of the proposed adjustment scheme, the results show that the proposed adjustment scheme can drive the process output closer to the process target and reduce the total quality loss significantly.