EAGER: Cybermanufacturing: Predictive Analytics Models and Techniques for Intelligent Cybermanufacturing

Georgia Tech Research Corporation, Atlanta GA

Investigators

Abstract

Enabling information infrastructure is the major element of the emerging service-oriented cybermanufacturing paradigm. Quality and cost are the major issues that 3D printing service providers (i.e. manufacturers) are facing as they are the critical measures for keeping their services competitive and affordable. The bidding price from a manufacturer or 3D printing service provider for a submitted job is crucial in market competition. This EArly-concept Grant for Exploratory Research (EAGER) project is the development of three data analytics approaches in a pilot prototype system to provide accurate cost estimation of 3D printing jobs. They are shape mining, user-product usage based analytics, and a hybrid approach combining shape mining with statistical mining of user-oriented data for enhanced cost prediction. The data analytics models and service-oriented framework will provide a generic and systematic approach to improve the efficiency of next-generation cybermanufacturing infrastructure, enabling manufacturers to make timely and sound decisions based on evidence and insight derived from deep learning from big data. This data science-based approach will enable better understanding of the power and the potential of cybermanufacturing in manufacturing efficiency enhancement and decision making.