Title: Using a neural-fuzzy approach for improving the perpendicular magnetic recording head manufacturing process
Authors: A.K.K. Fok, H.C.W. Lau, Y.K. Tse
Addresses: SAE Magnetics (H.K.) Ltd., 6 Science Park East Avenue, HK Science Park, Hong Kong. ' Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hunghom, Hong Kong. ' Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hunghom, Hong Kong
Abstract: Magnetic write width targeting is a critical process in achieving high process yield and good quality in manufacturing perpendicular magnetic recording heads for hard disk drives. The current approach is an experimentally based feedback process in which the quality is highly dependent on the testing accuracy and the sensitivity assumptions. This paper proposes a new approach based on neural-fuzzy based modelling for improving the quality of the magnetic write width targeting. The model consists of a neural network for: 1) suggesting changes in the lapping parameters, together with a fuzzy reasoning mechanism; 2) obtaining fine-tuned lapping parameter values based on the parameters derived from the neural network. Through improvement in the magnetic write width targeting process, the overall process yield, and product quality can be improved and the overall cost of the magnetic recording heads will be significantly reduced.
Keywords: neural-fuzzy modelling; areal density; product quality; inventory control; process control; neural networks; fuzzy logic; perpendicular recording heads; magnetic recording heads; magnetic recording head manufacturing; magnetic write width targeting; hard disk drives; HDDs; process yield.
International Journal of Intelligent Information and Database Systems, 2010 Vol.4 No.5, pp.413 - 430
Received: 17 Feb 2009
Accepted: 13 Aug 2009
Published online: 03 Oct 2010 *