Data Science and Predicting Material Performance in Additive Manufacturing (Carleton Lab)

Hybrid (both Remote and On Site)

The U.S. Army is working in conjunction with Columbia University to build material models that predict the performance of additively manufactured materials. The research team will apply data analysis and machine learning (ML) techniques to establish a relationship between raw material feedstock, additive manufacturing process parameters, and the performance of the additively manufactured (AM) bulk material. The research will consider characteristics from the nano- through to the macro-scale.
The research will progress in phases with an end goal of ML models predicting material performance in a multitude of environments. We are looking for students to
1. Perform a broad survey of open source additive manufacturing data and wrangle the most robust and readily available data for analysis of a single metal/metal alloy
2. Structure and build a query-directed material database
3. Build traditional data models that identify nano-scale, micro-scale, and/or macro-scale raw material feedstock characteristics critical to the quality and performance of the printed, bulk AM material
4. Build traditional data models that identify manufacturing process parameters critical to the quality and macro-scale bulk AM material performance
5. Build machine learning models that predict the micro- and macro-scale quality and performance of additively manufactured materials
Ultimately, the program will print and perform a bevy of material tests in Columbia’s Carleton Laboratory to verify this initial data analysis work. These data models, in conjunction with the laboratory tests, will lay the foundation for ultimately develop the constitutive equations for the bulk, additively manufactured metal/metal alloy.

If possible per COVID limitations, you will be exposed to multiple projects, including on-site testing and qualification of real-life materials.

Position Dates: 6/1/2021 - 8/31/2021

Direct Supervisor: William Hunnicutt & Adrian Brügger

Hours per week: 35

Paid: Yes

Credit: Yes

Number of positions: 3

Qualifications: AI programming and development and/or continuum mechanics of materials

Eligibility: Junior, Senior, Master's (SEAS only)

Adrian Brügger, [email protected]

If possible per COVID limitations, you will be exposed to multiple projects, including on-site testing and qualification of real-life materials.