A new technique of cotton bale management using clustering algorithm has been proposed. The method is based on the grouping cotton bales of similar kind into respective categories using k-mean square clustering algorithm. A set of 500 cotton bales were clustered into 5 categories by minimizing the total within-group squared Euclidean distance around the 5 centroids. In order to cluster bales of different categories, 8 fibre properties, viz., strength, elongation, upper half mean length, length uniformity, short fibre content, micronaire, reflectance and yellowness of each bale have been considered. Once the bales are clustered into different categories, it is possible to prepare consistent bale mix for consecutive laydowns on the basis of frequency relative picking method.
[1]
Richard A. Johnson,et al.
Applied Multivariate Statistical Analysis
,
1983
.
[2]
Yehia E. El Mogahzy,et al.
Theory and Practice of Cotton Fiber Selection
,
1995
.
[3]
Bok Choon Kang,et al.
A simplified optimization in cotton bale selection and laydown
,
2000
.
[4]
Yehia E. El Mogahzy,et al.
A Statistical Approach for Determining the Technological Value of Cotton Using HVI Fiber Properties
,
1990
.
[5]
Anja Vogler,et al.
An Introduction to Multivariate Statistical Analysis
,
2004
.