摘要

Because of more and more complexity of an operation environment in today's industrial production process, it is difficult to monitor the process operation quality and to estimate the performance efficiency based on the existing mathematical model and knowledge. This paper proposes a new method to estimate the process capability, and a new criterion for capability and performance assessment. This method is based on kernel function and fuzzy analysis hierarchy process (FAHP), which can improve the adaptation of process capability analysis. The device process capability can be estimated by FAHP with main variables, which are determined by interpretive structure modeling. The estimators of these indices overcome uncertainties caused by data fluctuation in the traditional process capability, and could strongly improve the robustness and adaptability of the process capability estimation and diagnosis. The proposed methods are used in a simulation of the Tennessee Eastman process. The results demonstrate the efficiency and validity of the presented approach. The proposed method can provide more performance decision information of industrial process to help decision makers evaluate and diagnose the state of the production devices, and improve the process operations.