2026 ASEE Annual Conference & Exposition

This abstract for a full paper addresses embedding artificial intelligence (AI) algorithms into an engineering computational modeling course as a continuation of previous work introducing AI concepts into engineering education [1]. While initial efforts focused primarily on generative AI and machine learning as a means of building early AI literacy, this study expands the scope by integrating practical algorithmic techniques directly into the coursework of a second-year engineering course. The goal was to improve students’ comprehension of AI methods, increase their awareness of ethical and practical implications, and evaluate their motivation to seek out further information about AI.

In this implementation, MATLAB™ toolboxes for neural networks and fuzzy logic were introduced alongside conventional modeling approaches. Students applied these AI methods to practical engineering problems, including image processing tasks such as object recognition. One exercise compared the effectiveness of AI-driven image detection techniques with more traditional approaches, such as FOR loops and threshold-based image processing, for identifying an apple. This comparison allowed students to directly evaluate the trade-offs between classical programming and AI-based solutions, highlighting where AI algorithms can provide greater adaptability and efficiency, and where simpler methods may be more appropriate.

To assess student outcomes, a survey instrument was administered both before and after the AI learning modules. The pre-survey established baseline levels of comprehension, awareness of limitations, and motivation to learn more about AI, while the post-survey expanded this scope to include measures of ethical understanding, recognition of bias, and evaluation of AI outputs. Results revealed increases in students’ self-reported comprehension of AI concepts and their ability to articulate the limitations of fuzzy logic and neural networks in terms of training requirements and data dependency, as well as an increased awareness of ethical implications and potential biases. However, consistent with findings from prior research [1], students’ overall motivation to independently seek additional information about AI topics dropped slightly.

The study suggests that embedding AI algorithms within an engineering computational modeling class provides students with a more practical understanding of how machine learning tools operate. By situating AI methods in direct contrast with conventional approaches, students not only gained technical knowledge but also developed an understanding of when such tools should be applied. The paper concludes with a discussion of survey findings, anecdotal student understanding of the algorithms, and implications for expanding AI integration across the engineering curriculum.