2026 ASEE Annual Conference & Exposition

The recent rise of AI tools in society is being echoed in the classroom, with ChatGPT and other LLM tools becoming increasingly prominent in both student and educator usage. However, this is not all that ‘AI’ means in the wider engineering context, and students graduating into industry roles will need understanding of the various ways that machine learning and artificial intelligence are being deployed in engineering companies.

This paper explores the development and expansion of AI/ML tools now being used in industry, with a focus on those embedded in engineering simulation software and electronic design automation (EDA) solutions. This includes not only generative AI tools but also more ‘traditional’ machine learning approaches such as neural networks, meta modeling and more. We look at how these tools and their underpinning fundamentals can be taught within the engineering curriculum, highlighting industry software which already contains AI and new coursework resources to leverage that.

Two products that are useful for Design of Experiments (DoE) are looked at in this paper: Ansys optiSLang, process integration and design optimization software and Synopsys Design Space Optimization AI (DSO.ai) software. Essentially, the AI/ML part of these tools generate fast prediction models from automated DoE workflows and validates the generated models with further simulation – an extremely powerful tool in industry, and a good example of scientifically applied machine learning for students.

In the long term, engineering curricula will likely change to include specific AI/ML-focused courses. In the shorter term, however, it will be necessary to expose students to these concepts within other courses. A typical engineering optimization course, taught in the third or fourth year of an undergraduate degree, might currently lead students through the mathematical formulation of optimization problems, introduce the Euler-Lagrange equations and then focus on the evaluation of different algorithms based on computational complexity and suitability for the specific problem. This could be a natural point at which to also introduce industry tools such as Ansys optiSLang software which contain another layer of modeling complexity, using machine learning to essentially carry out this algorithm trade-off evaluation within the software. Synopsys DSO.ai software, as EDA product, can teach similar optimization foundational topics in a more Integrated Circuit (IC) design focused curriculum.

This paper discusses (1) curriculum resources to easily embed Ansys optiSLang software into an existing course, as a lab or homework exercise and (2) a case example of integration of Synopsys DSO.ai software at Purdue University. Authors hope these examples show how industry can support deeper curriculum integration of AI/ML as a topic.