Abstract

Choosing between combined control charts and separate charts depends on three factors: experience, the extensiveness of out-of-control signals, and management objectives. These recommendations apply to multiple-tool and multiple-shift situations. If experience is that sawtooth patterns in a combined chart cannot be eliminated, then it is better to use separate charts. If the combined chart uncovers special causes that affect the two tools or shifts differently, then use separate charts. The extensiveness of rules to interpret the charts should affect which type of chart is used. More extensive rules usually mean quicker indications of out-of-control conditions. But this also increases the likelihood of false alarms. It also is important to know if management prefers detection of a single condition or multiple conditions being out of control. In general, separate charts are better for finding special causes that are peculiar to one tool or shift.