Visualization of machine clustering for a Taiwanese IC packaging foundry

Authors:
Hsu-Hao Yang;He-Yau Kang;Tzu-Chiang Liu
Affiliations:
Institute of Production Systems Engineering and Management, National Chinyi Institute of Technology, 35, Lane 215, Chungshan Road, Taiping 411, Taiwan;Department of Industrial Engineering and Management, National Chinyi Institute of Technology, Taiping 411, Taiwan;Department of Industrial Engineering and Management, National Chinyi Institute of Technology, Taiping 411, Taiwan
Venue:
Expert Systems with Applications: An International Journal
Year:
2007

Citing 6
Cited 0

Algorithms for clustering data

Algorithms for clustering data
Data clustering: a review

ACM Computing Surveys (CSUR)
Two-phase clustering process for outliers detection

Pattern Recognition Letters
Techniques of Cluster Algorithms in Data Mining

Data Mining and Knowledge Discovery
Integration of self-organizing feature map and K-means algorithm for market segmentation

Computers and Operations Research
Graph-Theoretical Methods for Detecting and Describing Gestalt Clusters

IEEE Transactions on Computers

Quantified Score

Hi-index	12.05

Visualization

Abstract

This paper applies a two-phase methodology to cluster 366 records of the wire bond machines for a Taiwanese IC packaging foundry, where six attributes of each machine are chosen to cluster. The purpose of clustering in this paper is to help the foundry more effectively assign a family of machines to orders that appear in various forms such as emergent or quality-demanding and are highly dynamic in nature. Given the clusters, we use the technique of parallel coordinates to plot each attribute's centers in clusters so that the foundry can take advantage of visualization to determine what machines can be assigned to orders. To plot these centers, we map the input values into the range of -1 and +1 so that proper parallel coordinates can be used. Originating from the graphs, we also compare the same machines appearing in clusters that are produced by applying different clustering methods. These identical machines form the important basis to support the clustering results for the management of machines.