Optimal process adjustment by integrating production data and design of experiments

This paper proposes a method to improve the process model estimation based on limited experimental data by making use of abundant production data and to achieve the optimal process adjustment based on the improved process model. The proposed method is called an Estimation-adjustment (EA) method. Furthermore, this paper proves three properties associated with the EA, which guarantee the feasibility and effectiveness of using EA for integrating production and experimental data for optimal process adjustment. Also, the paper develops a sequential hypothesis testing procedure for implementing the EA. The properties and implementation of the EA are demonstrated in a cotton spinning process.