On Statistical Design of the Cumulative Quantity Control Chart for Monitoring High Yield Processes

When dealing with high-yield processes, the cumulative quantity control chart (CQC chart) has been shown to be more suitable than the traditional c chart and u chart. However, the CQC chart shows an ARL-biased performance. In this article, we propose a design procedure to derive modified control limits for the CQC chart and the CQC chart with runs rules that can produce ARL-unbiased performance and maintain the predetermined false alarm rate. After extensive numerical work we have found that the proposed ARL-unbiased CQC chart is more sensitive than the conventional CQC chart in detecting process deteriorations.