Artificial intelligence optimization presents a number of advantages for mechanical engineers, together with sooner and extra correct designs and simulations, improved effectivity, lowered improvement prices via course of automation, and enhanced predictive upkeep and high quality management.
“When people think about mechanical engineering, they’re thinking about basic mechanical tools like hammers and … hardware like cars, robots, cranes, but mechanical engineering is very broad,” says Faez Ahmed, the Doherty Chair in Ocean Utilization and affiliate professor of mechanical engineering at MIT. “Within mechanical engineering, machine learning, AI, and optimization are playing a big role.”
In Ahmed’s course, 2.155/156 (AI and Machine Learning for Engineering Design), college students use instruments and strategies from synthetic intelligence and machine learning for mechanical engineering design, specializing in the creation of recent merchandise and addressing engineering design challenges.
Play video
Cat Trees to Motion Capture: AI and ML for Engineering Design
Video: MIT Department of Mechanical Engineering
“There’s a lot of reason for mechanical engineers to think about machine learning and AI to essentially expedite the design process,” says Lyle Regenwetter, a instructing assistant for the course and a PhD candidate in Ahmed’s Design Computation and Digital Engineering Lab (DeCoDE), the place analysis focuses on growing new machine learning and optimization strategies to review advanced engineering design issues.
First supplied in 2021, the category has rapidly turn out to be one of many Department of Mechanical Engineering (MechE)’s hottest non-core choices, attracting college students from departments throughout the Institute, together with mechanical and civil and environmental engineering, aeronautics and astronautics, the MIT Sloan School of Management, and nuclear and laptop science, together with cross-registered college students from Harvard University and different colleges.
The course, which is open to each undergraduate and graduate college students, focuses on the implementation of superior machine learning and optimization methods within the context of real-world mechanical design issues. From designing bike frames to metropolis grids, college students take part in contests associated to AI for bodily methods and sort out optimization challenges in a category atmosphere fueled by pleasant competitors.
Students are given problem issues and starter code that “gave a solution, but [not] the best solution …” explains Ilan Moyer, a graduate pupil in MechE. “Our task was to [determine], how can we do better?” Live leaderboards encourage college students to repeatedly refine their strategies.
Em Lauber, a system design and administration graduate pupil, says the method gave area to discover the appliance of what college students have been learning and the observe talent of “literally how to code it.”
The curriculum incorporates discussions on analysis papers, and college students additionally pursue hands-on workouts in machine learning tailor-made to particular engineering points together with robotics, plane, constructions, and metamaterials. For their closing undertaking, college students work collectively on a group undertaking that employs AI strategies for design on a fancy downside of their selection.
“It is wonderful to see the diverse breadth and high quality of class projects,” says Ahmed. “Student projects from this course often lead to research publications, and have even led to awards.” He cites the instance of a latest paper, titled “GenCAD-Self-Repairing,” that went on to win the American Society of Mechanical Engineers Systems Engineering, Information and Knowledge Management 2025 Best Paper Award.
“The best part about the final project was that it gave every student the opportunity to apply what they’ve learned in the class to an area that interests them a lot,” says Malia Smith, a graduate pupil in MechE. Her undertaking selected “markered motion captured data” and checked out predicting floor power for runners, an effort she referred to as “really gratifying” as a result of it labored so significantly better than anticipated.
Lauber took the framework of a “cat tree” design with completely different modules of poles, platforms, and ramps to create personalized options for particular person cat households, whereas Moyer created software program that’s designing a brand new kind of 3D printer structure.
“When you see machine learning in popular culture, it’s very abstracted, and you have the sense that there’s something very complicated going on,” says Moyer. “This class has opened the curtains.”