Change Management in Additive Manufacturing

AdditiveManufacturing AdditiveManufacturing

Background and Objectives

Current qualification procedures for production use of AM are labor and cost intensive. When a build parameter is changed there are not efficient methods to account for the changes and certify the resulting material performance. Methodology is needed to determine if a change in build parameters has an impact on material and product performance and verify as-built conditions fall within expected distribution based on process qualification. Sensor fusion will be used to build a machine learning model to generate predictive analytics in a metal SLM build. This methodology will fuse IR in-situ process parameters with resulting part quality, found by CT, to show expected results for future builds.

Supporting Agencies

AdditiveManufacturing AdditiveManufacturing

Contributors

Thrust 1: Logistic Regression based Classification to Predict Regional Anomalies for Voxelated Micro-CT Scans of Additively Manufactured Objects

References
  • A. Lang, J. Castle, D. A. Bristow, R. G. Landers, and V. S. S. Nadendla, "Logistic Regression Classification to Predict Regional Anomalies in Nominally Printed Volume of Separate Test Pieces," in Proceedings of the 33rd Annual International Solid Freeform Fabrication Symposium – An Additive Manufacturing Conference (SFF’22), 2022.