12 



BELL SYSTEM TECHNICAL JOURNAL 



per cent defective. These values have been determined by trial and error 

 and form the basis of the Ci d zones given in Fig. 7 of the Appendix. 



The appended DL tables (Double Sampling Lot Quality Protection) 

 provide a complete set of solutions using paired values of C\ and c^ deter- 

 mined from Fig. 7. These tables are constructed on the same principles 

 as the single sampling tables described above. 



Single Sampling — Average Quality Protection 



The solution for this plan considers the degree to which the entire in- 

 spection procedure screens out defects in the product submitted to the 

 inspector. Lots accepted by sample undergo a partial screening through 

 the elimination of defects found in samples. Lots that fail to be accepted 



1.0 1.5 2.0 2.5 3.0 



INCOMING QUALITY IN PER CENT DEFECTIVE 



Fig. 4 — Relationship between incoming quality, outgoing quality and AOQL 



by sample are completely cleared of defects. The overall result is some 

 average per cent defective in the product as it leaves the inspector, termed 

 the ''average outgoing quality," which depends on the level of per cent 

 defective for incoming product and the proportion of total defects that are 

 screened out. 



The solid curve of Fig. 4 shows how the average outgoing quality varies 

 for different values of incoming quality for a lot size of iV = 1000, a sample 

 size of w = 145 and an allowable defect number of c = 1. The curve is 

 based on the concept of incoming product of uniform quality treated 

 mathematically as an homogeneous universe. As the level of incoming 

 per cent defective gets higher and higher, more and more lots are completely 

 inspected. In turn, the average outgoing per cent defective increases, 

 reaches a maximum value (0.50%, in Fig. 4), and then falls off as a result 

 of rapid increase in the amount of screening. This maximum value is 

 termed the average outgoing quality limit (AOQL). 



