APPLICATION OF NEURAL NETWORKS AND THE TAGUCHI METHOD ON TOUCH COVER GLASS CUTTING QUALITY

NCUTIR > College of Management > Development of Industrial Engineering and Management > 【工業工程與管理系所】期刊論文 > Item 987654321/6204

Please use this identifier to cite or link to this item: http://ir.lib.ncut.edu.tw/handle/987654321/6204

Title:	APPLICATION OF NEURAL NETWORKS AND THE TAGUCHI METHOD ON TOUCH COVER GLASS CUTTING QUALITY
Authors:	洪永祥
Contributors:	工業工程與(工程)管理系
Date:	2013-08
Issue Date:	2017-09-30 13:22:00 (UTC+8)
Abstract:	In recent years, a touch glass product is ideal for touch screens and many other applications. The traditional cutting mode of touch glass product causes severe cracks and defects. This study developed a new cutting method, the heterotypic full-cut mode, which uses a toothed penett to make vertical cracks that penetrates deeper than the traditional scribing break. The vertical cracks simplified the fracture-free system. However, the heterotypic full-cut mode adopts a toothed penett, which does not require a breaking procedure, as its scribed glass automatically cracks. This paper applied neural network prediction model based on Taguchi method to discuss the optimal design of touch cover glass (CG) process parameters to improve product quality, appearance strength, and process yield, while reducing costs. This study experimentally validated the optimal value intervals and predicted CG strengths. Through Taguchi method, the parameter settings influencing CG strength were determined with a processing time reduced by 668 sec/pcs, productive capacity was increased by 4.4 times, and glass strength reached 600 Mpa. These results are significantly better than the traditional cutting methods. © 2013 ISSN 2185-2766. (14 refs)
Relation:	ICIC Express Letters Part B: Applications
Appears in Collections:	[Development of Industrial Engineering and Management] 【工業工程與管理系所】期刊論文

Files in This Item:

There are no files associated with this item.

View Licence

Loading...