Paper Info
Title
Demerit-Fuzzy Rating Mechanism And Monitoring Chart
Abstract
The relative magnitude of weights for defects has a substantial impact on the performance of attribute control charts. Apparently, the current demerit-chart approach is superior than the c-chart scheme, because it imposes different precise-weights on distinct types of nonconformities, enabling more severe defects to disclose the problems existing in the manufacturing or service processes. However, this crisp-weighting defect assignment, assuming defects are of equal degree of severity when classified into the same defect class, may be so subjective that it leads to the chart somewhat restricted in widespread applications. Since in many cases the severity of each defect is evaluated from practitioners' visual inspection on the key quality characteristics of products or services, when each defect is classified into one of several mutually-exclusive linguistic classes, a fuzzy-weighting defect assignment that represents a degree of seriousness of defects should be allotted in accordance. Therefore, in this paper a demerit-fuzzy rating mechanism and monitoring chart is proposed. We first incorporate a fuzzy-linguistic weight in response to the severe degree of defects. Then, we apply the resolution identity property in construction of fuzzy control limits, and further develop a new fuzzy ranking method in differentiation of the underlying process condition. Finally, the proposed fuzzy-demerit chart is elucidated by an application of TFT-LCD manufacturing processes for monitoring their LCD Mura-nonconformities conditions.
Year
DOI
Venue
2015
10.1142/S0218488515500178
INTERNATIONAL JOURNAL OF UNCERTAINTY FUZZINESS AND KNOWLEDGE-BASED SYSTEMS
Keywords
Field
DocType
Attribute control chart, demerit statistic, fuzzy number, LR-type fuzzy weight, fuzzy-number ranking, dominance degree
Visual inspection,Fuzzy logic,Control chart,Chart,Artificial intelligence,Fuzzy control system,Fuzzy number,Machine learning,Mathematics,Seriousness,Fuzzy ranking
Journal
Volume
Issue
ISSN
23
3
0218-4885
Citations 
PageRank 
References 
0
0.34
5
Authors
3
Name
Order
Citations
PageRank
Ming-Hung Shu121624.21
Jan-Yee Kung221.03
Bi-Min Hsu37810.18