By Karthika Swamy Cohen
Hybrid manufacturing systems, which employ additive or 3D-printing technologies along with subtractive methods that include more traditional machining techniques are becoming increasingly popular.
Modern 3D printing has been popular for rapid prototype development, but is now starting to make inroads into the manufacturing world. In additive manufacturing, 3D objects are constructed by successively depositing material in layers to create a predesigned shape.
Subtractive manufacturing starts with a block of material, from which material is successively cut away to obtain a given 3D shape. Subtractive manufacturing can be done by manually cutting the material but is most typically done with CNC machines that use computers to control machine tools, such as lathes, mills, and grinders.
3D printing generally involves various steps, starting with an input model, followed by orientation and positioning, addition of support structures, slicing, path planning, and machine instructions, which can all be enhanced by computational tools. The 3D model is first sliced into layers, which are then manufactured one at a time.
Behandish started with a 1988 quote by H. B. Voelcker, paraphrasing, “Our design tools aren't sophisticated enough for our manufacturing tools.” This remains true today.
Behandish presented examples of two of the latest software in subtractive and additive manufacturing, explaining how to expand the tools to go beyond what is available today.
The first tool he described, uFAB cloud based software, is an automated system for process planning for subtractive and additive manufacturing. It uses geometric reasoning and time and cost estimation to examine hundreds of millions of ways a 3D part can be made. It uses techniques from artificial intelligence planning at each step; each plan has different stages which allows material removal.
During the planning stage, the user removes as much material as he can, and if there is insufficient material, the software reports back alerting the uses that those particular features cannot be manufactured.
The next example Behandish described was a prop type software for 3D printing called MakeSim.
Here, mathematical morphology primarily computes the maximum subset of each slide that can be printed for minimum manufacturable feature size. Here, after the software computes different layers, it reports any parts that cannot be properly manufactured due to their size being smaller than the minimum feature slices.
Behandish then described DARPA’s TRANsformative DESign project whose aim is to design new tools to gain the level of complexity we need with design. The program aims to advance mathematics and computational tools to improve the complexity of design, and develop engineering tools to address design representation, analysis, and synthesis.
Behandish discussed an emerging paradigm of analytic methods that reevaluate a wide range of important geometric operations in measure-theory, convolution algebra, and digital signal processing.
These tools offer great promise for the use of computational design technology for traditional manufacturing, modern material structures, and fabrication processes.
Click here for more coverage of the 2017 SIAM Annual Meeting.