Plant Feature Detection and Performance Monitoring falls under the wider theme of Control and Performance Monitoring. Areas of investigation were multivariate data analysis, empirical modelling, feature extraction and process performance monitoring. Traditionally in process performance monitoring, univariate techniques have been applied to the data to assess the state of the process. However this approach fails to take account of the interactions between the variables where much of the information is contained. Through the application of multivariate statistical methodologies, an enhanced understanding of the process can be obtained. The project looked at the development and application of linear, non-linear, steady state and dynamic empirical techniques for the modelling and extraction of information from data collected on batch and continuous industrial processes. The methodologies also contribute to achieving manufacturing excellence.