Collaboration to reduce complexity of design in future advanced manufacturing

The Defense Advanced Research Projects Agency (DARPA; Arlington, VA) has chosen PARC, a Xerox company (Palo Alto, CA), to develop and deliver a new computational paradigm to remove the limitations that exist in current design platforms. The project aims to equip designers and engineers with new tools that integrate analysis and feedback on mission-critical performance criteria for material structures, geometric design complexity, and advanced manufacturing processes.

"With this work, we aim to disrupt part, assembly, and system design for both conventional and modern manufacturing processes. The computational innovations in modeling, planning, synthesis, and performance analysis will be developed to support the increasing freedom of manufacturing with unprecedented geometric and material complexity," says Saigopal Nelaturi, manager of PARC's Computational Automation for Systems Engineering area and lead on this project. "We aim to demonstrate the very first example of a fully automatic 'compilation' process from design requirements into fabrication instructions using the developed representations and algorithms."

"This project will radically change the field of computer-aided engineering (CAE) and how we design the next generation of complex engineered systems," says PARC CEO Tolga Kurtoglu. "FIELDS (Fabricating with Interoperable Engineering, pLanning, Design, and analysiS) will transform product design by creating a system that automatically searches high-dimensional spaces of shape, material, and process alternatives. It will generate feasible solutions to a set of design specifications, coupled with an ability to perform multi-fidelity, multi-physics-based analysis on the generated solutions. Significant enhancements in product performance and a shortened time-to-market are a few of the immediate benefits of the FIELDS technology. PARC has a unique innovation model, sitting at the center of academia, government, and commercial development, and we hope to make a big difference in advancing this vision."

As part of its FIELDS program, PARC aims to overcome limitations of existing computer-aided design technology, and enable robust and efficient synthesis and analysis of designs whose performance is tightly coupled to the manufacturing processes used to fabricate them. PARC, together with Intact Solutions (Madison, WI) and Oregon State University (OSU; Corvallis, OR), will develop and deliver a new computational design paradigm, engineered to extensively utilize high-performance computing capabilities.

"PARC and its partners will bridge the gaps between CAD, CAE, CAM, and new innovative manufacturing techniques. Today's fragmented approach has been holding digital manufacturing from enjoying what technology can provide today," says Ersin Uzun, vice president and director of PARC's System Sciences Lab. "The team is set to transform design and manufacturing by maintaining four in-depth views of an artifact (as-designed, as-planned, as-manufactured, as-analyzed) as it passes through the computational workflow from synthesis to fabrication. At each view, the structure of the physical artifact will be modeled representing shape, topology, and heterogeneous anisotropic material structure at six size scales. This computational framework will automatically provide manufacturability and performance feedback for synthesized designs, and will compile design requirements into a valid design with fabrication instructions. This paradigm will alleviate the burden on the designer to integrate computational and practical expertise from diverse disciplines, which is a significant bottleneck in today’s product lifecycle management systems. The framework will be adapted to specific manufacturing processes such as combined metal additive manufacturing and machining, and manufacturing with graded materials."

With DARPA and its partners, PARC's solution will incorporate:

1. New mathematical models, representations, and computations for physical artifacts with heterogeneous, anisotropic material structure;
2. Interoperable integration of synthesis, manufacturing planning, and analysis; and
3. Enormous design complexity by automatically searching very-high-dimensional spaces of shape, material, and process alternatives to help human designers discover physically realizable designs.
   

For more information, please visit www.parc.com.