(114b) Advances in Batch Data Analysis

Batch processes are ubiquitous in modern industry, namely in the production of high valued products and fine chemicals, due to their added flexibility and scalability when compared to continuous processing systems. Therefore, they are often considered as priority targets for data-driven process improvement activities. These activities may aim at different goals, such as improving process safety, efficiency, environmental impact, product quality and economical/financial bottom line results of the company. According to the nature of the activities, the tasks may need to be carried out online or offline, and always in a robust and manageable way - manageable in the sense that they can be handled within the knowledge and skills resources likely to be available in companies (possibly after a period of technology transfer). This may be particularly challenging to achieve for batch processes, due to the confluence of several well-known elements of complexity, such as: the tensorial nature of data collected from these processes, having a with a 3-way structure, X (I x J x K), where I is the number of batches, J is the number of variables and K is the number of measurements collected during a batch; their intrinsic dynamic and non-stationary nature; the occurrence of multiple stages, not all of them having equal importance in the final outcome; the existence of many variables involved, exhibiting a variety of typical time-profiles; among others.

In this tutorial, an overview will be provided about methodologies available to address data-driven process improvement of batch operations. Both online and offline activities will be targeted. A systematic way of organizing the different proposals available for conducting the analysis will also be provided, which may be instrumental for devising a roadmap for addressing the complexity of batch processes in more efficient and consistent way. The ultimate goal is fostering the exploitation of the information available from this type of processes in companies, by handling their complexity through the use of proper methodologies, within a clearly defined workflow.