Additive manufacturing or 3D printing has been hailed as the third industrial revolution because of the unique way it changes how products are conceived, designed, manufactured and distributed to end users. This research project addresses key scientific and engineering barriers that currently prevent the production of additively manufactured products at an industrially-relevant scale. In particular, the project focuses on how to better design products that will be produced via additive manufacturing by considering manufacturing constraints and challenges associated with assembling parts into final products. The study will result in fast algorithms for generating and selecting design concepts that have the minimum number of parts required to cover all functional requirements and that meet constraints imposed by the additive manufacturing process. This will facilitate high-volume, mass production of 3D printed products, thereby furthering the practicality and economic impact of additive manufacturing. The project spans several disciplines through a joint effort between the University at Buffalo (UB) and Rochester Institute of Technology (RIT). The project will enhance the curriculum related to design for additive manufacturing at the two participating universities by introducing results from this research into existing courses. Outreach activities will promote the recruitment of women and minorities to advanced manufacturing through workshops aimed at K-12 students organized in partnership with the Tech-Savvy program at UB. The research community will be engaged through a road-mapping workshop on computationally efficient methods in decision-based design.
The objective of this research is to define a prescribed theoretical framework and practical methods for classifying and generating physically integrated design concepts that can be realized effectively through additive manufacturing. The framework is based on the independence of functional requirements principle and modeling of physical integration problems as graph coloring problems. The research will result in new, computationally efficient (polynomial-time) graph-coloring algorithms based on discharging techniques in combination with Combinatorial Nullstellensatz. The new algorithms and design framework will be demonstrated and evaluated on representative design problems with the resulting designs being realized through additive manufacturing. Manufacturing productivity measures such as surface roughness, buildup time and cost, will be considered.
