A robust approach for root causes identification in machining processes using hybrid learning algorithm and engineering knowledge

Authors:
Shichang Du;Jun Lv;Lifeng Xi
Affiliations:
Department of Industrial Engineering and Logistics Engineering, School of Mechanical Engineering, Shanghai Jiaotong University, Shanghai, China 200240;School of Business, East China Normal University, Shanghai, China 200241;Department of Industrial Engineering and Logistics Engineering, School of Mechanical Engineering, Shanghai Jiaotong University, Shanghai, China 200240
Venue:
Journal of Intelligent Manufacturing
Year:
2012

Citing 7
Cited 1

The Strength of Weak Learnability

Machine Learning
Bagging predictors

Machine Learning
A neural network-based procedure for the monitoring of exponential mean

Computers and Industrial Engineering
Ensembling neural networks: many could be better than all

Artificial Intelligence
Neural Network Ensembles

IEEE Transactions on Pattern Analysis and Machine Intelligence
An ensemble of neural networks for weather forecasting

Neural Computing and Applications
Product lifecycle-oriented quality and productivity improvement based on stream of variation methodology

Computers in Industry

Identification of concurrent control chart patterns with singular spectrum analysis and learning vector quantization

Journal of Intelligent Manufacturing

Quantified Score

Hi-index	0.00

Visualization

Abstract

To improve product quality and productivity, one of the most critical factors for most manufacturers lies in quickly identifying root causes in machining process during ramp-up and production time. Though multivariate statistical process monitoring techniques using control charts have been successfully used to detect anomalies in machining processes, they cannot provide guidelines to identify and isolate root causes. One novel robust approach for root causes identification (RCI) in machining process using hybrid learning algorithm and engineering-driven rules is developed in this study. Firstly, off-line pattern match relationships between fixture fault patterns and part variation motion patterns are derived. Then, a hybrid learning algorithm is explored for identifying the part variation motion patterns. An unknown root cause is identified and isolated using the output of hybrid learning algorithm and engineering-driven rules. Finally, the data from the real-world cylinder head of engine machining processes are collected to validate the developed approach. The results indicate that the developed approach can perform effectively for identifying root causes of fixture in machining processes. All of the analysis from this study provides guidelines in developing root causes identification systems based on hybrid learning algorithm and engineering knowledge.