The design and engineering software market is in the middle of a major disruption on multiple fronts. New manufacturing techniques involving 3D printing, big-data, machine learning, and the scalable resources of the cloud are turning the industry on its head. Recently, a large emphasis has been placed on design optimization for additive manufacturing, with most of the focus on optimizing topology. The industry has coined this “Generative Design,” and it encapsulates all optimization techniques which can influence topology through physical simulation or parametrization.
The concept of Generative Design is compelling and liberating for designers and engineers. Exploring design spaces which could theoretically have never been explored due to constraints on time and creativity is intriguing to say the least. However, design conceptualization or idealization does not dictate a manufacturing process, it only ensures a design can be manufactured via a specific process. How the design should be manufactured given a specific process is an area that has seen little attention and certainly places limitations on Generative Design.
3D printing has an enormous degree of freedom for processing parameters. These parameters are typically set through “slicing” programs. For example, a slicer for FDM has many variables—layer width, number of layers for the outer shell, layer orientation, infill pattern, infill density, infill orientation, etc., all can be defined in a slicing program. The larger problem though, is that these parameters are currently set by a user manually, with no knowledge of how or why they should be defined. We see a large opportunity in helping users define these parameters through optimization techniques that couple manufacturing processes and functional simulation. We have coined this term “Intelligent Slicing Automation.”—the automation of slicing parameters through machine learning, optimization, and coupled physical, multi-scale simulations.