Data Analytics for Chemical Engineers (ChBE 6745)
This course will establish a foundation for handling data and understanding the difficulties that arise in trying to curate and learn from manufacturing systems data. It will introduce different classes of data that are commonly encountered such as time series and property data. It will introduce different classes of learning problems and the techniques for learning from the data. Students will learn basic programming skills in Python and be introduced to key libraries for machine learning and optimization. It will provide a map for students to connect the problem they are trying to address to the class of learning techniques that could be appropriate to use.
Data-driven Process Systems Engineering (ChBE 6746)
In chemical process engineering and manufacturing, decision-making requires the formulation and solution of complex optimization problems. This course first introduces students to the basics of optimization, such as formulations, linear, nonlinear and mixed-integer programming. The second and main part of the course focuses on case-studies where the optimization needs to rely on data (from experiments, historical databases or simulations). In this part, the course introduces key concepts with respect to sampling and design of experiments, data-driven modeling and regression, model validation and optimization using data-driven models. Data-driven decision making is discussed in the context of applications from the chemical manufacturing and processing industry, such as: process operation and control, process design, process and material design, multiscale process modeling, monitoring, planning, scheduling and supply chain optimization. Overall, by the end of the course the students will have a good understanding of how to couple data-analysis with optimization for decision-making in the chemical process industries.