An interactive and immersive human–computer interface for rapid composite part production design. (November 2018)

• Record Type: Journal Article
• Authors: McConnell, Roisin; Butterfield, Joe; Rafferty, Karen; Price, Mark; Murphy, Adrian; Burke, Robert; Legg, Adrian; Lemon, Richard
• Is Part Of: Proceedings of the Institution of Mechanical Engineers. Volume 232:Number 13(2018)
• Page Start: 2276
• Page End: 2285
• Publication Date: 2018-11
• Subjects: Assembly -- automated design -- knowledge assets -- composite tool design -- virtual design system; Mechanical engineering -- Periodicals; Engineering -- Management -- Periodicals; Manufacturing processes -- Periodicals; 629.8
• DOI: 10.1177/0954405417703427
• Languages: English
• ISSNs: 0954-4054

Abstract

This article addresses the need for better retention and exploitation of tacit knowledge for intelligent computer-aided design. It presents an automated design framework for the development of individual part forming tools for a composite stiffener incorporating parametrically developed design geometries. This work develops existing principles in knowledge-based engineering and parametric modelling beyond product design in the manufacturing planning domain. Outcomes demonstrate chronological benefits of automated process design methods as well as learning enhancements as the tacit knowledge data set can now include an applied element through an auto-generated virtual build environment. A virtual environment presenting a design concept to the planner for interactive assembly assessment was generated in twenty seconds and enabled the completion of virtual builds in support of the development of an optimal forming tool arrangement. This principle enables the addition of an experiential tacit knowledge feedback loop to further improve assembly planning for design concepts as they evolve. Challenges still exist in determining the level of reality required to provide an effective learning environment in the virtual world. Full representation of physical phenomena such as gravity, part clashes and the representation of standard build functions require further work to represent real physical phenomena robustly.