Design Like Nature
Amidst all of the confusion in terminology surrounding Rapid Technology and Additive Manufacturing (RTAM—itself a prime example of our semantic quagmire), “growing parts” remains the sole commonly shared metaphor, no matter how misleading. Whether we watch parts sparkle to life from a vat of primordial ooze beneath a dance of laser light or we excavate them carefully from the fine, elemental, cosmic dust, we collectively accept the idea that our creations grow into being, as though sentient forms of our own divine creation.
The choice of verb—to grow—alludes to the inherently biomimetic quality offered up by this nascent technology. Though the methodology has existed for as long as mankind has created devices (think Velcro), the term and an increased interest in biomimicry followed the book with that name, written by Janine Benyus in 2002. In it, she outlined instances where natural solutions offer insight into solving man-made challenges. The methodology has been warmly embraced by designers ever since, as it inspires us to look broadly for inspirations derived from millennia of field testing. Ironically, among the two most devoted followers of this line of thinking are those who fall into two disparate camps. On the one hand, there are the artists and industrial designers, inspired by the beauty and concinnity that results from balancing the man-made with the natural. On the other, there are the nuts-and-bolts engineers under DARPA or Department of Defense contracts, tapping into a long and sophisticated history of kill-or-be-killed Darwinian survival traits that have pushed tactical optimization to the extreme.
Optimized structure.
Since the birth of mass production, fabrication has evolved in an increasingly un-natural manner. Specifically, we’ve mastered the art of creating predictable, identical products to satisfy a market consisting of unique and varied individuals. “Identical” is a worse-than-useless attribute in Nature, as it signifies stasis, a lethal lack of genetic variation, or evolutionary stagnation. Moreover, parts made by mass-production methods are often sub-optimal, at least in part due to sacrifices demanded by the manufacturing processes that created them. Hollow-tube frame structures, for example, speak more to the ease of fabrication than to the unique mechanical demands of the situation. One glance across the skeletal or exo-skeletal structure of a living creature reminds us how unlikely it is that a uniform-walled, hollow tube actually meets any structural requirements in the most optimal manner. Yet we’ve come to accept that tradeoffs will be made in the interest of mass production and the demand for inexpensive functionality, and so the more idealistic sense of optimization has traditionally taken a back seat.
But when we’re offered the option to create a part molecule-by-molecule, the calculus changes. Suddenly we’re allowed to consider structures otherwise deemed prohibitively advanced. We can shift our focus from “real” to “ideal” and get away with it. In this situation, we may look to skeletal structures for enlightenment concerning our structural challenges. “Nature tends to feature non-linear phenomena, not amenable to straight lines and hard geometry. When you work in that environment, you must explore algorithms and methods that can generate the types of forms that will result”, according to Dr. Ken Trauner, orthopedic surgeon and mechanical engineer. “Truly natural structures rely on highly intricate forms not achievable through traditional means of manufacture”.
Huntsman Technologies has stepped into the biomimetic ring with its Conformal Lattice Structures. In short, this algorithm creates a lattice to strengthen a thin-walled product, resulting in forms resembling the trebecular structure found within a bird’s bone. The magic comes from the “conformal” part of the story, since the lattice actually generates itself based on surface normals at any given point, resulting in an especially efficient internal structure. This is something that can’t be achieved with traditional manufacturing methods.
But while this will undoubtedly create a strong and light structure, it falls short of a truly “nature-mimicking” solution. For that, modern 3D model-
ing packages tend to offer scripting opportunities, giving rise to architects and designers exploring the possibilities with Generative Design. This refers to a process where at least some of the geometry is created entirely by algorithm, based on input conditions, constraints, and intended results. Complex Voronoi structures, fractal bifurcation, and other forms of “natural patterning” can be scripted directly into the automated generation of the geometry. According to Wikipedia, “Generative design has been inspired by natural design processes whereby designs are developed as genetic variations through mutation and crossovers.” So now imagine that dual-walled, thin-wall part, populated with a structural lattice that was generated through recursive failure testing, based on the anticipated, simulated physical stresses. It’s structure where you want
it, and only as much as you really need. It boils down to CAD mimicking Darwin, enabled by the flexibility offered by freeform fabrication. It can be seen as a way to throw information and logic at a problem, instead of simply more material.
Nature refines its designs through a time-proven process: instantiate a variation of a successful design, perform extensive and destructive life-cycle testing, replicate surviving traits, repeat. It embodies the ‘fail early, fail often, fail fast’ method of design teaching. It begs for slight variations per instantiation and it defies old-school mass production thinking. And so when you grow that single, geometrically unique instance of your form, you quietly thumb your nose at mass production. Mass production hates uniqueness, while RTAM embraces and enables it. I tend to think that this represents the essential hallmark of the upcoming age of Rapid Manufacture. Certain products will be designed and optimized for their unique task, or to meet the specific needs of an individual user. And we can easily imagine a day when spare or replacement parts exist as data on servers ready to be fabricated into being, instead of stacked in warehouses filled with decaying inventory awaiting obsolescence and landfill.
Next Generation Thinking.
We have much to learn from nature, and some forward-focused universities are now embracing it as a discipline. But first, we must evolve a design culture wherein we learn to think as nature would, according to Tom McKeag, lecturer in Biomimicry at Berkeley and California College of the Arts: “Since our structures don’t change very much, they don’t adapt. Nature is all about adaptability, and that, of course, means real-time feedback loops and all the sensors, actuators and information programs that go with it. We are just catching on.”
Imagine the efficiencies to be found when we can 3D print heat exchangers that have the efficiencies of leaves. Or the capillary effect that allows tall trees to move water far beyond what logic would tell us is possible. As digital fabrication tools grow more robust in their capabilities and biomimicry becomes a lens for especially adventurous designers, we may well see a stunning new array of products and materials come alive before us.
But the rewards for this convergence of RTAM and Biomimicry may be found in boundless new fields yet to be defined. Andrew Hessel, co-chair of Bioinformatics and Biotech at Singularity University puts it this way: “The economics of 3D design and printing is compelling. Most of our manufactured goods are already made from a few source materials—metals, woods (just proteins), oil-based polymers, etc. The raw materials that comprise most cellular life, including human life, are plentiful and inexpensive. It is the organization of the materials, the information, that creates value. The evolution of new designs, perhaps the ultimate crossover in biomimicry, and facilitated by computers and computer-connected designers, could lead to incredible diversity and the personalization of virtually everything…Printing complex neurological structures could also be realistic—there are relatively few types of brain cells and electrodes are relatively simple, perhaps allowing brain-machine interfaces to be assembled from the ground up.”
To Andrew Hessel, this is already taking shape in the form of bio-printing—creating living tissue directly from design, using amino acids and proteins just as we currently spec polymers. “One breakthrough will come in the ability to ‘print’ multiple exotic materials simultaneously.” Except that “exotic” in my eyes is titanium or beryllium. To Mr. Hessel, it means muscle, bone and connective tissue.
