I-Corps: An Intelligent Re-Design Recommender System for Additive and Hybrid Manufacturing

The broader impact/commercial potential of this I-Corps project is the development of engineering software technology for additive manufacturing (AM). The proposed technology may be used to help solve some of the significant challenges by implementing a variety of intelligent assessments, re-design strategies, and learning processes. It provides a way to evaluate comprehensively the print-readiness of components and confidently identify which components may reliably be manufactured using AM processes.

Such evaluations may significantly reduce mistakes and printing failures, avoid costly trial-and-error exercises, and remove biases and inconsistencies that arise from manual evaluation processes. In addition, the proposed technology assists designers and manufacturers in rapidly converting problematic existing component geometries to successful 3D-printable designs, avoiding the need to design new components for AM fabrication from scratch.

The proposed technology presents a value for industries dealing with high-volume production by introducing intelligence and automation into the required assessments to save significant energy, computational, and financial resources.

This I-Corps project is based on the development of an engineering software technology integrating multiple intelligent assessment and re-design recommender systems for additive manufacturing (AM). While numerous industries are increasingly implementing AM, AM is not a feasible manufacturing method for many components, and improper implementation may cause resource waste and product failure.

The current technology presents a multi-variable algorithm for feasibility evaluation that monitors the design-readiness of components from several technical and economic perspectives before printing the part. Using computational approaches, this system intelligently calculates feasibility scores representing the precise feasibility status of various parts.

The proposed AM technology presents a fully automated correction system that rectifies the geometries of problematic components, allowing for immediate AM compatibility. The proposed re-design solutions also may provide successful geometries for 3D-printing components with minimal user-iteration and for different AM techniques. In addition, the proposed technology presents a design recommender system that intelligently analyzes an extensive database of closely related components for AM using machine learning techniques and provides effective re-design recommendations for component clusters whose current geometries are AM infeasible.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.