PVC(Polyvinyl Chloride) has been widely used to the daily commodities and medical appliances as one of ordinary four resins. PVC processes have unstable problems due to heat removal mechanism from exothermic reaction and strong nonlinear behaviors. As a result, most of batch process operations are mostly dependent upon the experience of the operators. As distributed control system(DCS) and plant information system(PIS) have been installed in industry, monitoring skills have been rapidly developed. The enormous amount of data being collected are summarized using one of data compression methods, MPCA(multiway PCA), which is capable of compressing the information down into low dimensional spaces. Process performance can be monitored by using statistical process control(SPC) charts constructed from MPCA model. It can be used to analyze assignable causes using contribution of the variables on the abnormal events. To develop the process monitoring model for industrial PVC reactor, in this study, MPCA model is introduced, and its control limits are determined from the in-control batch data, which are compared with real-time process measurement data to know the status of process. Furthermore analytical diagnosis methodology based on the contribution of out-of-control state are proposed and tested on PVC reactor. The proposed analysis tools show the efficient monitoring diagnostic results.