Model is usually necessary for the design of a control loop. Due to simpli.cation and unknown dynamics, model plant mismatch is inevitable in the control loop. In process monitoring, detection of the mismatch and evaluation of its in.uences are both demanded. In this paper we .rstly present several mismatch measures based on di.erent model descriptions. Then we categorize them into di.erent groups from di.erent perspectives and compare their potential in detection and diagnosis. A case study on a mixing process is presented and some remarks on related aspects are given.