Document Type
Article
Publication Date
1-2014
Publication Title
Journal of Computing and Information Science in Engineering
Volume
14
Issue
1
Publisher
American Society of Mechanical Engineers
Abstract
This paper presents an automated tool for estimating assembly times of products based on a three step process: connectivity graph generation from assembly mate information, structural complexity metric analysis of the graph, and application of the complexity metric vector to predictive artificial neural network models. The tool has been evaluated against different training set cases, suggesting that partially defined assembly models and training product variety are critical characteristics. Moreover, the tool is shown to be robust and insensitive to different modeling engineers. The tool has been implemented in a commercial CAD system and shown to yield results of within ±25% of predicted values. Additional extensions and experiments are recommended to improve the tool.
Recommended Citation
Please use publisher's recommended citation.
Comments
This manuscript has been published in the Journal of Computing and Information Science in Engineering. Please find the published version here (note that a subscription is necessary to access this version):
http://computingengineering.asmedigitalcollection.asme.org/article.aspx?articleid=1763546
Elsevier holds the copyright in this article