AI-Driven Generative Design: Overcoming Wicked Problems

Written by Jonathan Davis

Wicked problems, those that are complex, difficult to define and resistant to straightforward solutions are often associated with socioeconomic issues, from healthcare to urban planning.

Increasingly, modern developments in engineering and science have caused a paradigm shift in design from designing the possible to defining the possible.

The Sydney Opera House was delayed by this mentality, where the design proceeded a review of structural feasibility, instead focussing on delivering the requirements for a large, multipurpose and breathtaking concert hall. The feasibility was assumed and determined later.

On a smaller scale, but just as critically, component design for modern, complex, electronic and mechanical systems has become a wicked problem, where critical requirements must be met whilst ensuring feasibility and optimal performance.

The concept of generative design was developed to tackle this issue, as an iterative process where component design can be slowly adapted to meet a set of restraints. The iterative nature of this task almost predefines it as an algorithm and mirrors classical optimisation tasks, such as linear programming for profit maximisation.

Rapid developments in Generative AI have now made AI an obvious tool for supplementing, or replacing, the existing design process.

Not only can AI automate the repetitive and time-consuming design process that would otherwise need to be human-led, but it?can explore a vast design space a human may not even consider to create unconventional and innovative solutions.

At a time where there is fierce debate about the environmental impact of AI, this shows a real change for good. Innovative design can minimise material waste and maximise operational efficiency, transforming the way products and structures are conceived and developed.

At Applied Data Science Partners , we recognise, encourage and pursue the development of AI for improved operational efficiency and positive impact. As research and development of AI for generative design accelerates, we see no more important use case than satellite communications.

Modern satellites are at the centre of global technological infrastructure, advance the pursuit of scientific knowledge and can be used to save lives. This is why we are excited to be working with the European Space Agency - ESA on Disruptive SatCom Systems Design With Digital Generative Design.

Along with our partners, who bring specialist experience in SatCom, we will be researching and designing state-of-the-art AI systems that can produce 3D designs for satellite components.

We hope that this will help further the work of ESA outside of our atmosphere and encourage the utilisation of AI more widely across the aerospace industry.

Looking for more specialised consultancy? At ADSP we’re a team of data experts who build AI products with purpose.

We deliver data science projects for companies who want to harness the power that AI can bring to their organisation. Get in touch at [email protected].