(414a) Machine Learning for Engineering Course in Matlab and Python | AIChE

(414a) Machine Learning for Engineering Course in Matlab and Python


Hedengren, J., Brigham Young University
Fry, A., Brigham Young University
Loftin, J., MathWorks
Wang, J., MathWorks
Perez, K., Chemical Industry
Last, S., Brigham Young University
Phillips, N., Brigham Young University
Burrell, J., Brigham Young University
Cook, P., Brigham Young University
Hanson, H., Brigham Young University
Jung, S., Brigham Young University
Larson, S., Brigham Young University
Crop, A., Brigham Young University
A surge in the chemical engineering industry interest in machine learning has generated a high demand for students who possess expertise in this domain. Many chemical engineering departments are looking for ways to offer elective machine learning courses or integrate machine learning into existing chemical engineering courses. Chemical engineering instructors encounter a challenge in curating readily available curriculum materials that cater to the engineering domain when offering a machine learning course. Other aspects to consider are the familiarity of instructors with programming, student programming knowledge, and choice of software. MATLAB and Python are among the popular tools for machine learning. MATLAB’s app-based workflows, self-paced online training courses, and Live Scripts lower the learning barriers for students and make it easy to introduce machine learning.

In this talk, we cover how the “Machine Learning for Engineers” course was translated from Python into MATLAB through the collaboration of MathWorks and Brigham Young University as part of the capstone project course and showcase translated modules. During this project, students were exposed to both programming languages, received help from domain and AI experts to apply machine learning to chemical engineering applications, and explored the use of ChatGPT for code translations. MATLAB modules developed in this project are freely available at https://github.com/APMonitor/mds on GitHub and File Exchange. These modules can be used as instructor-led course material or self-paced learning resources and they contain case studies related to additive manufacturing, concrete strength, lithium-ion batteries, steel plate defects, polymers, and thermophysical properties which are relevant to chemical engineers. Modules also cover fundamental machine learning concepts. By making these course modules available in MATLAB and Python, we aim to fill the gap in machine learning teaching resources in a platform of the instructor’s choice.