| About this Abstract |
| Meeting |
MS&T24: Materials Science & Technology
|
| Symposium
|
Standards for Data Science in Additive Manufacturing
|
| Presentation Title |
How Much Data is Enough Data in the Qualification of AM Parts? |
| Author(s) |
John Barnes, Kirk Rogers, Matt Crill, Wayne King, Kevin Slattery, Rick Russell, Eric Versluys |
| On-Site Speaker (Planned) |
John Barnes |
| Abstract Scope |
Qualification and certification require data. Foresight and careful planning can reduce testing and accelerate qualification. Data determines critical quality characteristics impacting the design and that the material/process can meet the design intent. For high consequence parts, the data is typically generated under a methodical and complex test program. Initially, data becomes a metric for pass/fail on acceptable parts when variations or excursions from the specified process happen. We will draw on our experience and describe a data specification to improve collection and efficiently enable materials and process specifications for qualification. Once qualified, changes to the process will drive the need for a re-qualification. Defining major and minor changes determines the “delta qual”. Lastly, we will suggest how the use of advanced techniques such as artificial intelligence and in situ monitoring could enable a reduction in the amount of data required for qualification without reducing confidence in the result. |