Giving Mechanical Engineers Intelligent Tools: A Project-Based AI Education Curriculum in Thermal Engineering

Synthesis: Li et al. (2026) propose a project-based AI education curriculum that integrates artificial intelligence into mechanical engineering at the University of Arkansas, with a particular emphasis on thermal problems and their interplay with electrical and computer engineering. Responding to the limited AI background, challenging traditional thermal system modeling, and heavy course loads common in mechanical engineering programs, the three-level curriculum (introductory, application, and advanced) embeds AI across existing thermal topics rather than adding standalone CS-oriented courses. The authors report that students, regardless of prior ML participation, showed strong enthusiasm for machine learning and recognized its importance to professional development, and that hands-on practice enhanced confidence and interest. The full syllabus, data, and code are released in open-access repositories.

Key Findings

1. Mechanical engineering students often lack training in electrical/computer engineering and programming; the curriculum's AI-integrated pathway addresses this gap with a reasonable number of class hours.
2. Traditional thermal system modeling relies on physical laws (conservation of energy, heat conduction, Navier-Stokes) that struggle to capture nonlinear, multiphysics, and multiscale behaviors — AI models offer a complement.
3. The curriculum is structured at introductory, application, and advanced levels, covering core and optional AI projects that emphasize thermal problems and multidisciplinary communication.
4. Prior work (e.g., vibration-signal ML integration) showed students exhibited strong enthusiasm for machine learning and recognized its critical importance to professional development, with hands-on practice significantly enhancing confidence and interest.
5. The complete curriculum — syllabus, data, and code — is publicly available in open-access repositories, supporting adoption by other institutions.

Curriculum Design & Approach

The proposed curriculum integrates AI directly into existing thermal engineering topics instead of requiring long-duration AI courses from computer science departments, which the authors argue would further increase academic workload. The three-tier structure (introductory, application, advanced) moves students from foundational AI models toward engineering-specific problem solving and multidisciplinary communication. This project-based approach mirrors Project-Based Learning principles, treating AI not as a standalone subject but as an engineering tool embedded in domain problems. The curriculum is available in open-access repositories, contributing to the broader movement of embedding computational and AI literacy in STEM Education and engineering programs.

What this means for practice

• Curriculum designers. Embed AI inside existing disciplinary topics rather than adding standalone CS-oriented ML courses — the four core projects sit within thermal problems in a 39.1-hour course, avoiding further load on mechanical engineering students.
• Curriculum designers. Keep the three-tier introductory, application, and advanced structure and scaffold the coding: the authors attribute student progress to worked examples and scaffolded instruction that reduced cognitive load.
• Instructors. Plan separate expectations for undergraduates and graduate students: undergraduates who met only minimal entry requirements produced lower-quality projects and grades than graduate students despite revamped projects and example code.
• Faculty developers. Treat educator capacity as part of the curriculum, and open the course with real-world comparisons of salary levels, career path diversity, and professional ceilings for mechanical engineers with AI expertise, the strategy the authors used to reduce students' psychological barriers to learning.
• Instructors. Adopt the released syllabus, data, and code from the open-access repositories as a starting point rather than building a thermal engineering ML curriculum from scratch.

Limitations

• Single-institution curriculum at the University of Arkansas, taught across Fall 2021 (trial year), Fall 2022, and Fall 2023, with no control or comparison group; the reported evidence is course scores and project outcomes.
• Very small numbers: 14 students completed the course at the end of Fall 2023, and the authors report lower enrollment in the Fall 2022 semester.
• Undergraduates and graduate students took the course together with different prerequisites; the authors state that undergraduates' grasp of core concepts and programming proficiency put them at a relative disadvantage that inevitably affected the depth and quality of their project work and grades.
• Outcome claims are prospective: the authors write that they hope to document curriculum changes in response to student feedback and test effects on engagement and grades, so impacts on retention, graduation, or job placement are not yet measured.

Connected Concepts

• Curriculum Design
• Engineering Education
• Project-Based Learning
• STEM Education
• AI in Education
• CS Education
• Computational Thinking
• Teaching
• Open Source

Connected Articles

• A review of intervention designs of LLM Integration in Undergraduate Computer Science Education — LLM Intervention Design in Undergraduate CS Education: A Scoping Review
• A principled way to think about AI in education: guidance for educators and policy makers based on goals, models — Principled AI Education
• Reshaping Undergraduate Computer Science Education in the Generative AI Era — Reshaping CS Education with Generative AI
• The AI disruption in engineering education: an analysis of changing student norms through cultural historical activity — AI Disruption in Engineering Education
• Using LLMs to Detect Growth in Computational Thinking in Introductory Physics — Using LLMs to Detect Growth in Computational Thinking in Introductory Physics

Citation

Li, C., Hu, H., Dunlap, C., House, N., & Wai, J. (2026). Giving mechanical engineers intelligent tools: A project-based AI education curriculum in thermal engineering.