Recognition of Control Chart Patterns using Discriminant Analysis of Shape Features Monark Bag

Each control chart pattern (CCP) has its own geometric shape and various related features can represent this shape. The shape features can represent the main characteristics of the original data in a condensed form. Different patterns can, therefore, be efficiently discriminated based on these shape features extracted from the control chart plot. In this paper, a feature-based heuristics approach is proposed that can recognize nine main types of CCPs, including the mixture pattern. The important shape features are identified and extracted, and then, the heuristics in the form of a decision tree is developed based on discriminant analysis of the extracted shape features.