In a previous paper [9], the authors presented an improved p chart which provides a large improvement
over the usual p chart for attributes. However, this control chart can detect large increases in the
nonconforming rate p but is not efficient for detecting small increments of the process parameters.
For situations like this, control charts with double sampling (DS) have two possible advantages over a
single sampling plan. First, it may reduce the total amount of inspection and second has the advantage
of giving a lot a second chance before taking a decision.
In this paper, authors apply a double sampling procedure for the improved np chart and provided
the designs that minimize the ARL1 against increases of 50%, 100% and 200% in the process
non-conforming rate. This procedure was applied to processes with in-control non-conforming fraction
p0 equal to 0.005, 0.010 and 0.020 and for ARL0 ≥200 and for ARL0 ≥ 370.4. The optimal designs
were given for average sample sizes such that np0 = 0.5, 1.0, 2.0 and 4.0. In all these cases it was
found that the double sampling minimizes the number of samples needed until the detection of one
item out of control in the process.
In a previous paper [9], the authors presented an improved p chart which provides a large improvementover the usual p chart for attributes. However, this control chart can detect large increases in thenonconforming rate p but is not efficient for detecting small increments of the process parameters.For situations like this, control charts with double sampling (DS) have two possible advantages over asingle sampling plan. First, it may reduce the total amount of inspection and second has the advantageof giving a lot a second chance before taking a decision.In this paper, authors apply a double sampling procedure for the improved np chart and providedthe designs that minimize the ARL1 against increases of 50%, 100% and 200% in the processnon-conforming rate. This procedure was applied to processes with in-control non-conforming fractionp0 equal to 0.005, 0.010 and 0.020 and for ARL0 ≥200 and for ARL0 ≥ 370.4. The optimal designswere given for average sample sizes such that np0 = 0.5, 1.0, 2.0 and 4.0. In all these cases it wasfound that the double sampling minimizes the number of samples needed until the detection of oneitem out of control in the process.
การแปล กรุณารอสักครู่..
