TRONOX's Bunbury pigment plant is developing a data analytics system using PI Server Asset Framework and PI Vision to automatically detect when the plant and equipment performance deviates from expected standards.The system compares actual plant performance as represented by data in the plant's PI System data historian. It uses a combination of Theoretical and Relational models and Standard Specifications coded into PI Server Framework to recognise deviations and generate PI Server Event Frames. The PI Server Event Frames are used to drive role-specific Alert lists in PI Vision for review, action and acknowledgement.The presentation will discuss:• The context for the development • The features of the system and an example of how the method can be used to improve plant OEE. • Issues encountered in the development