To extract the greatest value from advanced manufacturing, a shift to codesign with AI is needed to realize the full potential of DfAM (design for additive manufacturing). 3D printing offers unlimited design freedom, giving the advantage to reduce weight via intelligent lattices and topology optimization that were not previously possible with subtractive methods. AI codesign can iterate designs in significantly less time than a human engineer alone. The cybersecurity of these processes must be taken into consideration due to the fact that these methods are currently suited for extreme operating environments that benefit the most from these manufacturing breakthroughs. If a part is going through AI design workflows it is most likely a critical part and its failure would be catastrophic. This can be seen in automotive applications with Bugatti’s additively manufactured brake caliper.
A failure of a brake caliper due to cyber-sabotage in AI codesign could easily cause a fatal car crash. The weight savings of 40% sprung weight cannot be ignored when considering fuel consumption, and as automotive technology trickles down into the consumer market, these AI optimizations will contribute towards driving down carbon emissions. It will be critical to already have in place the detection capabilities to determine if AI codesign has been altered either via tampering with operator inputs into performance characteristics or via switching out a less than optimal iteration into the production environment. BISON will provide valuable insights into post-failure analysis should an incident occur due to AI codesign components. If you are interested in a deeper dive into benefits the automotive industry can gain from these design approaches please read this blog post from nTop – the leading software company focused on these methods.
As you can see (above) the controls to create a repeatable workflow are highly complex and alterations here would be hard to notice and very likely make it to the manufacturing floor. BISON gives you the ability to upload a trusted source manufacturing file, in this scenario, a validated AI codesigned bracket. Using Bison’s ability to hash an as-manufactured-component against the trusted file an operator can quickly learn that alterations have been made even if the components look visually identical.
The Department of Defense is already pursuing these methods in flight critical components. Episode #43 of The Cool Parts Show showcases a helicopter heat exchanger that has enhanced performance over traditionally machined versions.
“Advanced Engineering Solutions applied geometry that could only be made through additive manufacturing to the redesign of a heat exchanger for the gearbox oil of a helicopter. The result: four times the cooling in a heat exchanger one half the size of the original. Gyroid lattices inside this heat exchanger maximize inner surface area to achieve the more effective heat transfer.” (The Cool Parts Show Episode 43)
Does your manufacturing process depend on mission critical AI co-designed components? If so, please contact us at firstname.lastname@example.org to learn more about how BISON can secure your AM environment.