AI-Driven Computational Design
This course delves into cutting-edge AI and machine-learning methods to develop AI-based designs of objects and physical experiments, produce manufacturing workflows, and convert digital designs into manufacturing instructions.
Understand how to develop an intelligent design and manufacturing workflow by integrating cutting-edge AI and machine learning techniques.
Reduce manufacturing lead time through AI-driven industrial processes.
Increase workforce productivity by introducing advanced digital design and manufacturing techniques.
Reduce R&D costs by introducing AI/machine learning methods to convert performance-driven designs into manufacturable designs.
Secure a competitive edge by enhancing technological efficiency, design accuracy, and manufacturing speed.
Create three-dimensional objects employing generative design techniques.
Throughout this course, you will
- Understand how to develop an intelligent design and manufacturing workflow by integrating cutting-edge AI and machine learning techniques.
- Learn how to represent parametric/procedural designs, and explore generative AI methods to represent design spaces (e.g., design families).
- Understand how to convert digital designs to manufacturing instructions.
- Learn how to predict design performance using virtual testing, numerical simulation, and AI surrogate methods.
- Delve into performance-driven design workflow and principles of generative and inverse design.
- Learn how to use machine learning for the design of physical experiments and design optimization.
- Understand how to develop an intelligent design and manufacturing workflow by integrating cutting-edge AI and machine learning techniques.
- Learn the practical applications of state-of-the-art AI tools such as Fusion 360, OpenSCAD, ChatGPT, Google Colab, and Adobe Firefly.
- Recognize the capabilities and limitations of current advanced manufacturing hardware.
This course is aimed at
- Design and engineering professionals seeking to modernize product design and optimize manufacturing processes through AI-generated solutions.
- Manufacturing and production specialists looking to optimize production, enhance operations, improve resource utilization, and reduce costs.
- Product I+D experts looking to accelerate innovation cycles, create more personalized products, and stay ahead in competitive markets.
- Professionals from relevant fields that encompass consumer products, medical devices, electronics, architecture, and defense, among others.
Certificate
Meet your instructor
Wojciech Matusik
Professor at the Department of Electrical Engineering and Computer Science
The MIT Learning Experience
Learn through practice
Practice processes and methods through simulations, evaluations, case studies, and tools
Learn from others
Connect with an international community of professionals while working on projects based on real examples
Learn on demand
Access all online content and watch videos from anywhere
Reflect and apply
Bring the newly learned skills into your company through examples of technical work environments and good thoughtful prompts.
Show your success
Earn a Professional Certificate and 2.4 Continuing Education Units from MIT xPRO (MIT xPRO CEUs).
Learn from the best
Gain insights from MIT faculty and industry experts, with one live session every two weeks.
What is the course comprised of?
This course follows intricately crafted modules that address developing intelligent design and manufacturing workflows.
Module 1 – Introduction to Computational Design
-
Design Representation
- High-Level Design Representations.
- Geometric Representation.
- Designs as Programs.
- Assignment: Converting a Design into a Digital Representation
Module 2 – Design Spaces
- Parametric Parametric Modeling
- Design Grammars
- Geometric Deformation Methods
- Assignment: Constructing a Design Space
Module 3 – Generative AI for Learning Design Spaces
- Overview of Generative Methods
- Linear Models
- Non-linear Dimensionality Reduction
- GANs
- Diffusion Models
- Large Language Models
- Assignment: Building a Generative Model
Module 4 – Mapping Design to Performance Metrics
- Design vs. Performance
-
Examples of Engineering Simulation Methods
- Rigid/Articulated Body Simulation
- Finite Element Analysis
- Computational Fluid Dynamics
-
Building AI Surrogate Models
- Physics-Informed Neural Networks
- Performance Evaluation using Large Language Models
- Assignment 1: Building an AI Surrogate Model
- Assignment 2: Predicting Performance Using ChatGPT
Module 5 – Design Optimization (Inverse Design)
- Introduction to Inverse Design
-
Examples of Inverse Design
- Shape Optimization
- Layout Optimization
- Topology Optimization
- Design Optimization using Large Language Models
-
AI Methods for Design Optimization with Limited Number of Experiments
- Bayesian Optimization
- Assignment: Design Optimization Using Bayesian Optimization
Module 6 – Generative AI for Creative Design
-
Generative models for images and video
- Diffusion Models
- Subjective Performance Evaluation
- User Experience Design Examples
- Future Directions for Generative AI in Creative Design
- Assignment: Hands-on Experiments with Diffusion Models
Module 7 – Generative AI for Digital Manufacturing
- Translating Designs to Manufacturing Instructions
- Process Optimization
- Bridging the Gap between Digital Design and Reality
- Assignment: Constructing the Translation Process for a Digital Manufacturing Process