(506d) Problem-Based Learning on Incorporation of Data Analysis Skills into a Senior Course

Through years of production, chemical process industry (CPI) accumulated rich data asset. In the age of information and AI, processing facilities are using data in new ways to improve efficiency, reliability, and safety. However, skilled workforce is in shortage. To embrace the digital transformation, chemical engineering students need to get in-depth training on data analytics skills.

Monte Carlo methods defined broadly a statistical approach to provide approximate solutions to mathematically complex optimization or simulation problems facing uncertainty by using random sequences of numbers. Monte Carlo simulations are used to model the probability of different outcomes in a process that cannot easily be predicted due to the intervention of random variables. It is a technique used to understand the impact of risk and uncertainty in prediction and forecasting models. Monte Carlo methods have been widely used in chemical industry. In the application of asset health, heat exchanges’ performance was modeled considering manufacturing tolerances and uncertain flow distribution.

In a “Advanced Analysis” class, the fundamentals of statistics, outlier detection, and Monte Carlo Simulation were explained. Then a real-world problem was assigned per the request from the industry. The students were divided into groups of 4-5 people each, and they were asked to figure out which heat exchangers in a refinery need to be cleaned based on condition rather than time, with the help of Monte Carlo Simulation and statistics analysis [2]. This approach was well-received by the students, some of them eventually became process engineers or maintenance engineers, who are using this skill in their work.