Using Process Modelling as a Job-Aid to Reduce Composites Manufacturing Risk
Many current composites manufacturing practices still rely on experience, engineering judgement and methods of trial-and-error. However, as the size and complexity of the composites end-product, and production rates increase, these approaches are no longer efficient or practical. Attempts to address this important issue have focused on three key research thrusts: automation, artificial intelligence (eg. machine learning and data analytics), and simulation(eg. process modelling and Integrated Computational Materials Engineering (ICME)). This paper focuses on the use of process modelling as an enabling technology.Whilst the composites industry does recognize the value of using process modelling, there still are many perceived barriers that limit its use. Key concerns include: having the ‘right-sized’ tools for a given problem, and knowing how to use these tools. Additionally, most commercially available process modelling software packages are developed for use in the late stages of program development and production. However, the ability to make effective manufacturing decisions at this late stage is limited. The work presented in this paper is part of a broader initiative to establish a knowledge framework to effectively manage composites manufacturing risk. Two thermal management case studies are presented to show how to manage uncertainty in both experiments and modelling, and how we can make better manufacturing decisions even with partially reliable test and analysis data.
Author: Janna Fabris
Conference: SAMPE Seattle 2017