Big Data Analytics and Fundamental Modeling I

Tools for data collection and analysis on a massive scale have been driving force in development of Big Data analytics in many fields. Data-driven approaches such as data-mining, multivariate analysis, machine learning etc. allow for robust models and are aided by fast parallel computational tools such as Hadoop and Spark. However, unlike many others domains, big data in chemical and process engineering is embedded with rich structure due to fundamental principles that govern processes. The traditional use of first-principles based modeling has led to proliferation of control and optimization techniques which have been extremely successful in safe & optimal plant operations. This session encourages submissions which demonstrate the application of big-data analytics and first principles in process industries. We especially encourage submissions discussing amalgamation of big-data analytics techniques with fundamental knowledge, traditional tools, models, and algorithms. Examples of such applications include: supplement model development with data mining and pre-processing tools, application of fast statistics and visualization to process live data; hybrid data-driven models employed in model predictive control; Internet-of-things approach to process plants and supply chains; machine-learning models for fault diagnosis; innovative application of machine learning to chemical engineering.