Towards data-driven additive manufacturing processes

Abstract

Additive Manufacturing (AM), or 3D printing, is a potential game-changer in medical and aerospatial sectors, among others. AM enables rapid prototyping (allowing development/manufacturing of advanced components in a matter of days), weight reduction, mass customization, and on-demand manufacturing to reduce inventory costs. At present, though, AM has been showcased in many pilot studies but has not reached broad industrial application. Online monitoring and data-driven decision-making are needed to go beyond existing offline and manual approaches. We aim at advancing the state-of-the-art by introducing the STRATA framework. While providing APIs tailored to AM printing processes, STRATA leverages common processing paradigms such as stream processing and key-value stores, enabling both scalable analysis and portability. As we show with a real-world use case, STRATA can support online analysis with sub-second latency for custom data pipelines monitoring several processes in parallel.