ABSTRACT
Similar equations can be used to prepare control charts for the number of defects per unit and the number of defects. If uis considered to be the average number of defects per unit in the complete set of tests, then it can be calculated as the total number of defects in all samples divided by the total number of units in all samples. Then, the central control line is it and the control limits are
it ± 3[~J~ If the parameter of interest is the number of defects per sample, C, then the central control line becomes the total number of defects in all samples divided by the total number of samples. The control limits are
C ± 3[cf/2 One example of the benefits of employing control chart techniques as a part
of a quality control effort is demonstrated by the experience of a cutting tool manufacturer [4]. New York Twist Drill is a manufacturer of tight-tolerance twist drill bits for the aerospace, automotive and general metal working industries. Prior to the serious application of statistical process control methods, New York Twist Drill was experiencing scrap rates of about 13 %and a rework rate of about 15 %. Following a top-down commitment to the statistical process control philosophy, quality control techniques were implemented that were based upon in-shop control charts. These changes resulted in the machine operator's having a sense of ownership and increased pride in the manufacturing processes which provided almost immediate cost savings. Scrap was reduced to less than 1% and rework costs were reduced by about 75 %. In addition, each step in the manufacturing process became easier to accomplish due to the cumulative benefit of having improved consistency and quality occurring throughout each of the preceding stages.
