Digital Twins and AI for Predictive Analytics

Catalog description

This course explores the transformative power of digital twins in harnessing data-driven insights and improving decision making using predictive analytics. In this course, students will learn principles and applications at the dynamic intersection of digital twin technology, data analysis, and applied machine learning (ML) to support predictive analytics across engineering areas within research, industry, business, and government. The technical course content will progress data analysis, to inference, to applied ML in order to demonstrate the process of collecting, cleaning, interpreting, transforming, exploring, and analyzing data generated by digital twin models. Using this process, students will learn to extract pertinent information, communicate insights, and support decision making based on predictions of how engineered systems perform under future scenarios. The advantages of using visualization techniques to explore data and communicate outcomes will be highlighted throughout the course. This course will include lectures that embed concepts from across the AI broadly to support applied ML along the predictive analytics lifecycle. Select lectures providing additional breadth will include the implications of ethics and social justice in AI, how to select appropriate computing platforms for predictive analytics, and data management. By the end of the course, students are expected to be able to plan, design, and implement empirical research projects using statistical, computational, and quantitative applied ML techniques, predict system response to support data-driven decision-making using digital twins, and discuss the ethics of AI-driven decision making such as bias.