The CPLS-based method provided a complete monitoring strategy for input and quality variables, however, no literature considered the influence of missing measurement in this method.

We derive the conditional distribution of scores and residuals given the observable measurement, and employ this probabilistic measurement to construct monitoring statistics.

We perform probabilistic analysis on monitoring statistics in the presence of missing measurement, and then derive the uncertain ranges of monitoring statistics caused by missing data.

The proposed method is illustrated by its application in Tennessee-Eastman process.