An algorithm is presented for designing np control charts on the basis of their statistical performance. The algorithm gives the sample size and control limits for an np chart (the 3-triplet np chart) based on the specified values of (a) the in-control fraction nonconforming value p0, a downward shifted value p−, and an upward shifted value p+; (b) the corresponding design run lengths (RL); and (c) the corresponding probabilities (W). Specifically, whenever the process fraction nonconforming equals one of the three specified values, the 3-triplet np chart will produce a signal in a predetermined run length with a predetermined probability. This feature makes the operating characteristics of the np chart highly specifiable and provides the quality assurance engineers with better perception and more control in the design and operation of the np charts. The 3-triplet np chart is designed by seeking an optimal combination of the sample size and the control limits.
