The Bionic Partition is the world’s largest metal 3D-printed airplane component. A partition is a dividing wall between the seating area and galley of a plane, and it is a challenging component to design because it must be very thin and strong.
The new Bionic Partition—created through a pioneering combination of generative design, 3D printing, and advanced materials—is almost 50% lighter than current designs, and it is also stronger. This weight savings translates to fuel savings and carbon reduction. The final design illustrates a novel use of “bio computation,” and it demonstrates an ultrahigh-performance result beyond typical engineering rules of thumb. The Bionic Partition is currently undergoing 16G structural testing as part of the process for certification and integration into the current fleet of A320 planes. When slime mold grows, it creates a complex network that is both efficient and redundant. It is efficient because it connects a given set of dots (food) with a minimal amount of lines. And it is redundant because each dot touches at least two lines—so if any line is removed, the dots remain connected in the network. We developed a “biological algorithm” based on slime mold to link critical connection points in an airplane partition. Then we ran a process of artificial evolution in the computer called “generative design” that generates, evaluates, and evolves tens of thousands of design options.
Combining our custom techniques of data science and bio computation, we can derive results that are both highperforming and unexpected. The process is not about achieving cold-blooded efficiency. Rather it is about expanding our creativity. The Bionic Partition is pushing the limits of several technologies, but it is on track for a real-world industry application. When applied to all A320 planes flying today, this new design approach could save up to one million tons of carbon emissions per year. And this is just for one airplane part. The next step is to apply it to other parts, and eventually the whole airplane.