Number of Defects - Control Chart
Sometimes the attribute of concern is the number of defects (c) in an item rather than the number of defective items in a sample: for example, the number of flaws per square metre of cloth. In this situation the concept of counting non-defects, for comparison with the number of defects found, is meaningless. Such a process typically conforms to the Poisson distribution, of mean value ¯c and standard deviation √ (¯c) and the control limits are determined as:
Warning limits = ¯c ± 1.96 √ (¯c)
Action limits = ¯c ± 3.09 √ (¯c)