Generative design (CAD/CAE)?
Generative design is a powerful new approach that utilises artificial intelligence (AI) and machine learning (ML). Instead of an engineer drawing the initial part, the engineer simply tells the computer the rules. These rules include specific goals and limitations, such as the material to be used, how heavy the final part can be, the manufacturing process, and how well it needs to perform. The AI then automatically creates a huge variety of possible designs that all meet these specified criteria. Essentially, traditional CAD starts with a design and checks it against the rules, but generative design starts with the problem parameters and uses computing power to discover the absolute best solution.
Scope of Generative Design:
Generative design has a broad and transformative reach across several major sectors and is redefining how engineers approach product development and manufacturing. Generative design is widely adopted in industries where optimisation, performance, and complexity are critical factors:
- Aerospace: Used for designing extremely lightweight, high-performance structural components to maximise fuel efficiency and payload capacity.
- Automotive: Applied to reduce vehicle weight, optimise engine parts for thermal management, and design lighter chassis components.
- Architecture and Construction: Utilised for structural optimisation, facade design, and creating complex, efficient building systems.
- Industrial Machinery: Helps in designing lighter and more robust equipment parts, reducing material costs and improving operational efficiency.
- Consumer Goods: Enables the creation of personalised, customised, and ergonomically optimised products.
Core Applications:
The core purpose of using generative design is to create more efficient products and make manufacturing simpler. A main application is lightweighting components. This means significantly cutting down on the part’s weight while keeping its required strength or even making it stronger. Generative design also allows for part consolidation, where several separate pieces that were traditionally assembled can be merged into one complex, unified part. This reduces the steps needed for assembly and lowers the chances of a part failing. The technology is used for performance optimisation, such as refining a design to improve a car’s aerodynamics or enhancing thermal management for electronics. Generative design makes mass customisation possible, allowing companies to efficiently produce unique, personalised products based on specific user needs or data.
Real-life examples of Generative Design:
- Airbus A320 Partition Wall: Airbus used generative design to create a bionic partition wall (separating the cabin from the galley) that was 45% lighter than its traditional predecessor, saving significant fuel and reducing emissions.
- General Motors Seat Bracket: GM utilised the technology to develop a seat bracket that was 40% lighter and 20% stronger than the original component.
- WHILL Electric Wheelchair: A Japanese manufacturer used generative design to reduce the weight of a battery case for a portable motorised wheelchair by 40%, decreasing the overall cost by consolidating parts.
- New Balance Midsoles: The footwear company partnered with Formlabs to create customised, performance-optimised 3D-printed shoe midsoles tailored to specific runner data.
Benefits of Generative Design:
Generative design offers significant advantages by making the entire process faster and more efficient. It accelerates design cycles by automating the time-consuming and repetitive work. The AI can explore thousands of potential options quickly, which drastically speeds up the time it takes to get a product ready for the market. This process leads to optimised performance because the resulting designs are stronger, lighter, and more efficient in terms of material use, while being perfectly tuned for specific goals like dissipating heat or having maximum structural integrity. It boosts innovation and creativity by producing novel, often organic shapes and unconventional layouts that human engineers might never have considered. By only using the material necessary for strength, it achieves cost and material efficiency, minimising waste and offering substantial savings and sustainability benefits. It also allows for the creation of complex geometries that are typically only practical or even possible to manufacture using advanced methods like additive manufacturing (3D printing).
Software with AI-generative design:
Listed below are some of the popular software with AI-generative design.
- Autodesk Fusion 360
- PTC Creo Generative Design
- Siemens NX
- MSC Apex Generative Design by Hexagon
Drawbacks and Challenges:
Despite its benefits, implementing generative design comes with several hurdles. There are high upfront costs and expertise requirements. This technology demands significant computational power and requires staff with specialised knowledge in AI and machine learning, posing a barrier for smaller companies. Additionally, human oversight remains essential. The AI works strictly within the rules given to it and lacks human intuition or subjective judgment regarding things like aesthetics or cultural context. Engineers must always step in to check the generated options for practical feasibility and safety compliance in the real world. The process can also lead to a paradox of choice because the software generates a massive number of valid design solutions. If the original problem was not perfectly defined, this sheer quantity of options can be overwhelming to sort through. There are also potential bias concerns; AI algorithms might unintentionally adopt biases found in the data they were trained on, which could negatively impact the diversity and inclusivity of the final design outcomes. Finally, engineers used to traditional CAD face a steep learning curve, and organisations must manage the disruption caused by shifting established workflows to accommodate the new technology.
Conclusion:
Generative design is fundamentally reshaping engineering by using AI and machine learning to find optimal solutions based on an engineer’s constraints. This approach drives significant benefits, including accelerated design cycles, profound lightweighting, and the creation of innovative, high-performance parts across industries like automotive and aerospace. Real-world success stories, such as the lighter Airbus partition wall, prove its value. Adoption is slowed by challenges like the high upfront costs, the need for specialised human oversight to evaluate the many options, and the steep learning curve for engineers. Despite these hurdles, generative design is set to become a standard tool, unlocking efficiencies and design complexity previously unimaginable through traditional CAD methods.
References:
https://www.ptc.com/en/technologies/cad/generative-design
