Machine Learning Methods for Abnormality Detection in Hard Disk Drive Assembly Process: Bi-LSTM, Wavelet-CNN and SVM

2018 ◽  
Author(s):  
Masayuti Simongyi ◽  
Prabhas Chongstitvatana
Keyword(s):  
Machine Learning ◽  
Hard Disk Drive ◽  
Hard Disk ◽  
Disk Drive ◽  
Assembly Process ◽  
Abnormality Detection ◽  
Learning Methods ◽  
Machine Learning Methods

Fourier-assisted machine learning of hard disk drive access time models

2013 ◽  
Author(s):  
Adam Crume ◽  
Carlos Maltzahn ◽  
Lee Ward ◽  
Thomas Kroeger ◽  
Matthew Curry ◽  
...  
Keyword(s):  
Machine Learning ◽  
Hard Disk Drive ◽  
Hard Disk ◽  
Disk Drive ◽  
Access Time

scholarly journals PENINGKATAN PRODUKTIVITAS PADA LINE ASSEMBLY STOPPER CRASH STOP DENGAN LINE BALANCING

2016 ◽  
Vol 5 (1) ◽  
Author(s):  
Qomaratun Nurlaila
Keyword(s):  
Hard Disk Drive ◽  
Assembly Line ◽  
Hard Disk ◽  
Disk Drive ◽  
Work Load ◽  
Line Balancing ◽  
Assembly Process ◽  
Main Root ◽  
Vibration Machine ◽  
Rubber Material

PT. XYZ is manufacturing company, it produces HDD (Hard Disk Drive) components with plastics and rubber material. One of HDD products which used rubber material is Stopper Crash Stop, it’s produced by assembly process. The actual productivity from July – September 2013 indication that have average productivity is less than the standard. The actual productivity is 8,949 Pcs/man/day, but the standard productivity is 10,350 Pcs/man/day. The Main Root Causes, which contribute to productivity of assembly line of Stopper Crash Stop less than the standard are workplace that more than range of operator’s hand, time of machines operation are different, work load are not balanced. And to increase the productivity of assembly line Stopper Crash Stop to achieve the standard done by  Kaizen concept in phases and sustainable. 1st kaizen is layout kaizen, re-layout table of vibration machine by make sit position of operator A and operator B is equal. 2nd kaizen is machine kaizen, standardize time of machines operation refers to the fasters one. 3rd kaizen is layout kaizen, optimize wide of table for vibration machine. And 4th kaizen is line balancing kaizen, make work load of operator is balance. The Productivity before kaizen is 8,949 Pcs/man/day and the productivity after kaizen is 12,323 Pcs/man/day (the productivity increases 38%). 

scholarly journals Feature Selection to Improve Performance of Yield Prediction in Hard Disk Drive Manufacturing

2020 ◽  
pp. 420-428
Author(s):  
Anusara Hirunyawanakul ◽  
◽  
Nuntawut Kaoungku ◽  
Nittaya Kerdprasop ◽  
Kittisak Kerdprasop
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Problem Solving ◽  
Hard Disk Drive ◽  
Information Gain ◽  
Hard Disk ◽  
Disk Drive ◽  
Absolute Error ◽  
Support Vector ◽  
Yield Prediction

Hard Disk Drive (HDD) manufacturing is one real-world application area that machine learning has been extensively adopted for problem solving. However, most problem solving activities in HDD industry tackle on failure root-cause analysis task. Machine learning is rarely applied in a task of yield prediction. This research presents the application of machine learning and statistical techniques to select appropriate features to be used in yield prediction for the HDD manufacturing process. The seven well-known algorithms are used in the feature selection step. These algorithms are decision tree (C5 and CART), Support Vector Machine (SVM), stepwise regression, Genetic Algorithm (GA), chi-square and information gain. The two prominent learning algorithms, Multiple Linear Regression (MLR) and Artificial Neural Networks (ANN), are used in the yield prediction modeling step. Yield prediction performance has been assessed based on the two evaluation metrics: Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). Yield prediction with MLR shows higher accuracy than yield estimation traditionally performed by human engineers. Resulting to conclusion that the proposed novel learning steps can help HDD process engineers to predict yield with the better performance, especially on applying GA as feature selection tool, the MAE is reduced from 0.014 (yield estimated by human engineer) to 0.0059 (yield predicted by MLR). That means error reduction is about 60%.

Sign in / Sign up

Export Citation Format

Share Document