Concept exploration is a major part of the product development process.
Right from the start, engineers study requirements and goals to create a concept that meets them. Then, a series of dedicated design, analysis, and testing iterations are performed to further optimize and achieve a final, fabricated production-ready outcome. However, these iterations take time – with often more of it needed for increasingly perfected products.
With ever more computer processing power, CAD suites have consistently expanded the tools they offer engineers and product designers to support their process. And today, with the onset of AI and machine-learning algorithms, an alternative, computational approach to design emerges to present the field with many advantages.
Computational design fundamentally changes the way engineers interact with computers as design tools.
In this new approach, product designers shift their focus from designing the outcome to structuring the procedure of designing the outcome. They develop pre-defined sets of rules, criteria, and specific instructions from requirements for a computer to execute, calculate, and render as a unique proposal, its data, and geometry.
In computational design, exploring information is handled by the system, which frees valuable time and resources. And generative design moves this approach forward even more.
As an application of computational design, generative design leverages the power of AI and machine-learning to boost the concept exploration phase. Product designers develop goals, rules and constraints, and the system powers through the evaluation and iteration of possible outcomes that meet them. It draws from previous results and designer feedback to propose a ranked set of alternatives that that can be further refined towards a final product.
Generative design is an exploration catalyst. It helps engineers evaluate more possibilities in less time to achieve optimal outcomes faster, with better insights, and desired performance.
It is a collaborative approach that augments engineers’ capabilities and creativity with computational force. It can lead to innovative designs free from typical geometric and manufacturing biases. And it can help optimize important resource allocation for a more efficient product development process.
Product development starts with a thorough analysis of mission and product requirements. The results become a specific set of goals with rules, considerations, constraints, relationships, and more that are then explored through the design process to achieve an outcome that is further tested, analyzed, and refined.
Incorporating generative design happens in the transition from requirements to goals to include the precise framework that the product development team will explore through a computational solution. It is the designers’ role to first envision and develop the concept and specifications from which to design, then to guide and supervise the system’s execution of them into proposals that feed the development process.
The computer renders a multitude of alternatives from the designed goals and all available data and experience. These are analyzed and ranked, then evolved, explored, and rendered as a new iteration of more perfected alternatives. The designer participates in each step, guiding the system towards a desired proposal. And the process can be repeated until a final selection is made, which can then be further developed into an optimal product result.
While generative design may add an extra layer of abstraction in the early design stages, its applications in product development have the potential to help engineers streamline the process for better, faster results.
In aerospace engineering, weight-to-performance is crucial.
One of the ways generative design applications present an advantage is through innovative topology optimization solutions that use fewer resources in products that outperform heavier, traditionally designed alternatives. NASA JPL prototypes have already employed the novel design approach for future concept structures, as have different aircraft designers and manufacturers in their own solutions.
Below is an example from a component design in one of Newton’s recently developed products.
The line draft illustrates the initial concept designed through conventional methods, while the rest of the sequence shows its topology evolve through generative design. With this approach, we reduced 53% of the original concept’s weight while maintaining all other properties and positive margins within our required resistance and safety factors.
Development time is another key metric in which generative design shows promise through its augmented concept-to-product validation process.
As exploration is handled by a computational suite, it can generate a multitude of alternatives to help identify the best solutions to a design requirement in record time. It has the potential to speed up development times for specialized components and products. And it can help engineers drive aerospace innovation further with optimized development schedules and cost.
At Newton, we’re committed to deliver reliable, cutting-edge engineering services that help further science, discovery, and the future of aerospace.
Because of this, generative design is now a part of our product development capabilities designed to help you, and your mission, reach beyond. If you’d like to learn more or are interested in exploring what we could build together, please get in touch.