Manufacturing is a highly complex field involving materials, machines, and processes that interact in ways we often don’t fully understand. This complexity poses a significant challenge when we attempt to optimize manufacturing processes, introduce new materials, or enhance efficiency. While artificial intelligence (AI) has shown great promise in helping us identify patterns, make predictions, and drive innovation in manufacturing, understanding how some AI models make their predictions or decisions is very difficult. This “black box” problem is a significant challenge, especially when manufacturers need to trust AI tools to improve production, quality, and safety.
The PRISM team aims to develop methods and tools for interpreting and explaining how powerful AI models make decisions in manufacturing. By seeing what’s going on inside the “black box”, we can uncover new scientific or engineering insights. By doing this, we can identify fundamental pieces of knowledge—what we call “knowledge building blocks” (KBB)—that help us better understand how manufacturing systems work. These building blocks can also guide us in developing improved, physics-based models that more accurately reflect real-world manufacturing processes.
In this project, students will use real manufacturing data to build robust, data-driven AI models that represent key manufacturing processes and systems. We will also develop new techniques to “interrogate” these models, exploring and understanding how the AI reaches its conclusions. By uncovering how the models work, students will help extract the essential insights that form the foundation of our understanding of manufacturing.
The PRISM team is committed to sharing and publishing the tools and methods we create so that other colleges, universities, and industry partners can benefit. By making our work widely available, we hope to expand the number and types of manufacturing systems that can be interpreted and understood using AI.
This research is ongoing, and the scope and skills needed are expected to evolve from year to year. Students joining the team in Winter 2026 will work towards the following objectives:
- Conduct intensive background research into various classes of manufacturing systems for which high impact will be gained if interpretability can be achieved.
- Develop a comprehensive understanding of the key AI models and frameworks.
- Develop expert knowledge of our manufacturing data sets and begin developing the processes and tools to “unpack” the foundational patterns identified by different AI models. Eventually, this understanding will be codified into more generalized “manufacturing knowledge building blocks” (MKBB).
- Demonstrate initial understanding of the AI models targeted for interpretation research.
In addition to learning about the technical details of the project, we plan to schedule at least one visit to an automotive manufacturing plant or operation (advanced lab, research facility, etc.)
PRISM subteams are designed to be flexible, encouraging creativity, student growth, and collaborative research. Students may move between subteams based on workload and individual interests, allowing for varied experiences and personal development. As students gain skills and confidence, they can expect to take on greater responsibilities and work across multiple aspects of the project.
Each subteam is guided by an experienced student leader who works closely with the PRISM faculty PI and supports team members in reaching project goals. Students will collaborate closely with their subteam leader to ensure progress and learning.
Applications are welcome from first-year undergraduates through master’s students. The team seeks highly motivated, self-driven individuals who are interested in contributing to innovative, team-based research. The most dedicated students will be encouraged to remain with the project for more than the two-semester minimum, with opportunities to grow into student leadership roles as they gain knowledge and experience.
Students with relevant interests and skills are encouraged to apply. While the project operates through subteams, applications are for the project as a whole, not for specific roles or subteams.
Data Science Fundamentals (3-5 students):
Develop workflows, interpretations, and processing methods for the use of data to explain manufacturing phenomena. Knowledge and interest in mathematical, statistical, and data science fundamentals. Development of new characterization algorithms and techniques.
Preferred Skills: Advanced data science, linear algebra, statistical analysis, strong Python programming skills
Likely Majors: DATA (Grad), MATH, STATS
Tool Development (3-5 students):
Develop full-stack tools tailored to managing the manufacturing data and supporting an infrastructure for “plug and play” evaluations.
Preferred Skills: Good full-stack programming skills, particularly database design and management for HPC (high-performance computing), with an understanding of data science. Python and AWS tools.
Likely Majors: CS, CSE, DATA
Model Validation (3-5 students):
Implementation of a wide variety of data techniques, including AI/ML.
Preferred Skills: React.js, Python, MATLAB, HPC experience
Likely Majors: DATA
Manufacturing (2-3 students):
Practical knowledge and experience of manufacturing theory and process. Interest in analytical and practical investigation of manufacturing problems.
Preferred Skills: Must have basic Python programming skills and data skills.
Likely Majors: ENGR, MFG (or any with practical experience)
Operations & Project Management (2 students):
Project management, technical writing and documentation, coordinate team activities/build team cohesion, facilitate communication between subteams
Preferred skills: Ability to lead and manage teams, good organizational and communication skills (written & oral), relationship building.
Likely Majors: BBA, Any STEM (with relevant experience)
Apperentice Researchers (3 Students)
Interest in project material and a willingness to develop skills. Students will be integrated into the operations of a subteam. Reserved for first- and second-year undergraduate students.
Likely majors: CS and any STEM student with good math skills.
Jeffrey Abell, Ph.D., P.E.
Professor of Practice in the Mechanical Engineering Department. Dr. Abell comes to the University of Michigan after more than 30 years at General Motors, where he most recently served as Director and Chief Scientist for Global Manufacturing. At GM, Dr. Abell led global research in vehicle electrification, lightweight materials processing, automation, and advanced manufacturing systems. He developed several industry-first battery manufacturing technologies, managed multi-million dollar Department of Energy research projects, and is a three-time recipient of the GM “Boss” Kettering Award, the company’s highest technical recognition. Dr. Abell has managed global teams and research collaborations, including international assignments in Portugal and Germany. He is a licensed professional engineer and holds board and advisory roles with SME, ABET, and multiple engineering academic programs.
Team Logistics: The MDP-PRISM team meets weekly on Wednesday’s 5:00-7:00 pm, in person on North Campus. Led by the faculty PI, these meetings cover technical developments, team management, and strategy planning, and include hands-on working sessions addressing current challenges.
Subteams also meet in person each week at times that work best for their members. These sessions are focused on collaborative work, progressing towards semester goals, and addressing specific questions. The faculty PI may join subteam meetings as needed.
Subteam leaders will meet weekly with the faculty PI to regularly report the progress of each subteam. Since all students will be new to the project, subteam leaders will be chosen as members gain experience and demonstrate leadership skills. Experienced students aspiring to be sub-team leaders should express this in their MDP application. Students with a passion for research and a desire to develop leadership skills for team management have a unique opportunity to receive direct mentorship from the faculty PI.
The team will also maintain a Slack channel to coordinate progress asynchronously, seek support from others, document challenges, and generally keep each other informed.
Enrollment: Students will enroll in ENGR 255,355,455, or 599 for 2 credits per term. Enrollment is annual, with two-term commitments starting each January.
Course Substitutions: Honors, CS-ENGR/DS-ENGR/EE/CE-ENGR 355 and higher can count toward Flex Tech.
These substitutions/departmental courses are available for students in these respective majors. MDP does not yet have a formal agreement with other departments for substitutions/departmental courses not listed. Please reach out to your home department’s academic advisor about how you might apply MDP credits to your degree plan.
Citizenship Requirements: This project is open to all students on campus.
IP/NDA: Students who successfully match to this project team will be required to sign an Intellectual Property (IP) Agreement prior to starting participation in January.
Location: In-person participation is expected. Some team meetings and activities will be held remotely.
Summer Opportunity: Some Summer research internships may be available for qualifying students.
Learn more about the expectations for this type of MDP project