In general, attribute control charts use p, np, and u control charts when the data involved consists of two attributes.But when the attribute is more than one, the demerit control chart is present to overcome this problem.The assumption that must be met in this control chart is that the relationship between one sample and another is independent.In fact, this assumption is often not met, so that the correlation between one sample and another must be considered. One application of this case is in the industrial world, especially at PT Unitex Tbk on the New Ajl Dobby machine. In this paper, we will discuss the application of the Demerit control chart used to test the quality of the production process if in a process the number of defects each unit between observations is correlated from time to time.Then an effort is needed to obtain another control chart that is more suitable for the data, the control chart, which is appropriate and will be proposed in this paper, namely the Exponentially Weighted Moving Average (EWMA) control chart on demerit data. Merging the two graphs aims to increase the sensitivity of the control chart in detecting out of control. The results show that the EWMA - Demerit control chart will be more sensitive than the demerit control chart when the value is 0.1 is greater than or equal to lambda greater than or equal to 0.6.

Control Chart

Demerit system

Exponentially Weighted Moving Average