This thesis is concerned with an economic design of group control charts for a process with multiple streams. In the traditional group control charts, statistics are calculated using samples taken from all streams. In some situations where the cost of sampling and testing is high, however, it may be economical to take samples from a portion of the streams.
A modification of the group control charts is considered which chooses a portion of the streams at every sampling point and takes samples from the selected streams. For the proposed control charts, the expected cost per unit time is derived which involves the cost of false alarms, the cost of detecting and eliminating assignable causes, the cost associated with production in out-of-control state and the cost of sampling and testing. Optimum values of the number of streams selected, sample size and control limits are obtained by minimizing the expected cost per unit time. Numerical studies are performed to compare the proposed control charts with the traditional group control charts.
