Generative Design, Topology Optimization, 5 Recommendations
Huge change is coming to CAD - prepare for disruption!
Many different digital technologies are impacting industrial suppliers - in design, engineering, manufacturing, sales and marketing. How to Decide Which Technology to Keep Up With and Apply? is a tough question and no simple decision. For example, in their August 2018 webinar '6 Reasons your Product Design Won't Survive in 2020', Engineering.com cites 6 technology disruptors that will 'be bigger than CAD' - Generative Design, IOT, Augmented and Virtual Reality (AR/VR), Digital Twins, Products as a Service and Machine Learning.
We think those 6 and many others (for example, Additive Manufacturing (aka 3D Printing), eCommerce, Configurators (aka CPQ), AI, Adaptive Pricing) will impact industrial suppliers in the coming years. We intend to cover them all on this blog along with our recommendations for industrial suppliers. So far we've written about the impact of IIOT a number of times (for example IIOT Use is Accelerating Fast, Is Yours? and The Coming Impact of IIOT on Industrial Sales and Marketing) and 3D Printing (for example Allow 3D Printing for Prototyping but not Manufacturing and 3D Printing of Metal Parts is Now Mainstream!). Now, in this blog, we're focusing on Generative Design in general and its 'star' component, 'Topology Optimization' in particular!
What is Generative Design?
If you aren't already familiar with Generative Design the image above illustrates 4 iterations for an example design. Also see:
- Wikipedia's definition
- Autodesk describes the process as "mimicking nature's evolutionary approach to design" by iterating through successive design alternatives seeking topological design solutions for a given set of design goals and constraints including loading cases, boundary conditions, materials, manufacturing methods, and cost constraints
- CIMdata defines it as, "a process wherein the shape and composition of a product is determined using physics-based simulation or other analysis method that consider performance requirements and optimize objectives such as minimum cost or weight - the designer does not work directly with the materials or product models"
What is Topology Optimization?
As we said above Topology Optimization is perhaps the 'star' of Generative Design and is going to revolutionize design of products and components. Topology Optimization maximizes the performance of a design by positioning material optimally within a given design space, for a defined set of load cases, boundary conditions and constraints.
How Does Topology Optimization Work?
Simply speaking, Topology Optimization tools are basically algorithm driven FEA (or CFD) tools that vary material density, decreasing it in low stress areas, in multiple iterations to come up with optimal new designs - sometimes optimizing for cost or weight or other criteria. The key point is to produce design options based on functional goals and constraints. The resulting designs can look 'organic' and be difficult, if not impossible, to manufacture using traditional manufacturing techniques yet can be created with additive manufacturing (aka 3D Printing).
How will Generative Design/Topology Optimization Impact Industrial Suppliers?
For ~40 years Computer Aided Design (CAD) has been our key tool for designing, prototyping, manufacturing or building industrial products. As computing power and graphics improved and costs decreased CAD evolved in sophistication (2D, 3D, Solids, CAE, FEA, CFD, etc.) and integration with our business processes (PDM, PLM, CNC, ERP, CAM, CRM, etc.). Today CAD is ingrained in our design to manufacturing business processes and in many cases well beyond. For example, if your downstream maintenance process depends on exploded parts diagrams you probably use 3D CAD assembly models to create those diagrams or if your online marketing process offers 2D CAD drawings and/or 3D CAD models of your products as downloads to save your customers time these may be a byproduct of your CAD design process. Many of our business processes rely on data hand-offs to or from CAD (such as Prototyping to CAD or CAD to CAE).
CAD sophistication continues to evolve based on the availability of massive computing power as Generative Design, and Topology Optimization illustrate. For industrial suppliers they will enable many more design variants to be investigated and optimized far faster than a human designer could do so.
1. Recognize that Generative Design (including Topology Optimization) is still in its infancy. Expect rapid adoption in industries where product weight is critical, for example, in the aerospace industry or any racing vehicles. Decide whether this technology could be integrated into your design process and improve any of your products. Would a more 'organic' looking product be acceptable?
2. If so identify one of your products or components that could benefit from generative design and topology optimization then allocate a single designer or engineer to do a pilot design project to see what you learn.
3. Select a Generative Design/Topology Optimization software tool. Your existing CAD supplier/s may already offer these tools so we recommend that you start there with your own software research. Some focus on reaching a single topologically optimal solution, others on producing multiple design alternatives. Either way, you may find a new improved design or learn how to lighten your existing product. Carefully consider what the output of the generative design process is. If it is not a true CAD model that can be edited by your CAD system (for example to add fillets before 3D Printing) and used in your existing downstream processes such as those mentioned above then you can expect to change other processes too! Or, if the output isn't a CAD model how will you turn it into a CAD model?
4. Recognize that optimization leads to less resilience so a generative design may, but not necessarily, be less accepting of unexpected treatment or loading and hence there may be greater risk of failure. Therefore understand that you must be able to fully identify the functions, loading, constraints and boundary conditions because a resulting design will be optimal for those and perhaps those alone.
5. Re-analyze and possibly physically load test the resulting proposed design for resilience before finalizing it.
At CDS we monitor and share available research and new CAD technology as we continue to develop our industrial marketing technologies and services to help B2B industrial suppliers succeed with online marketing and eCommerce. As always, please share your comments below or, if you'd like our opinion on your use of CAD for marketing, please call us or click either button below:
* image credit to source companies that are linked to from each image