TIGHTENED INSPECTION PLANS-Sampling Plans AND OPERATING CHARACTERISTIC (OC) CURVES FOR AQL 6.50 DEFECTS PER HUNDRED UNITS REDUCED AND NORMAL INSPECTION PLANS-SAMPLING PLANS AND OPERATING CHARACTERISTIC (OC) CURVES FOR AQL=10.00 DEFECTS PER HUNDRED UNITS [Sampling plans-AQL=10.00] Comparable sampling plans Single. Double. 5 6 84 13 14 168 23 24 315 41 42 500 62 63 800 95 29 18 36 335 23 34 3385 46 516 64 65 TIGHTENED INSPECTION PLANS-SAMPLING PLANS AND OPERATING Characteristic (OC) CURVES FOR AQL = 10.00 Defects per HUNDRED UNITS [31 FR 4687, Mar. 19, 1966; 31 FR 4949, Mar. 25, 1966, as amended at 36 FR 18457, Sept. 15, 1971; Redesignated at 41 FR 42640, Sept. 28, 1976 and 42 FR 32514, June 27, 1977, and further redesignated at 45 FR 69424, Oct. 21, 1980 and 46 FR 63203, Dec. 31, 1981] § 42.141 Obtaining Operating Characteris tic (OC) curve information for skip lot sampling and inspection. The Operating Characteristic (OC) curve information (probability of acceptance) for skip lot sampling and inspection procedures described in § 42.121 is easily obtained from the single curve in § 42.142. The procedure for using the curve in § 42.142 is as follows: (a) Select any sampling plan for normal condition of container inspection from Table I or Table I-A of § 42.109. (b) For a given Quality of Submitted Lots, 100,, find the Percent of Lots Expected to be Accepted, Pa from the respective OC curve in § 42.140. (c) Refer to § 42.142 with this Percent of Lots Expected to be Accepted, Pa, and read the new Percent of Lots Expected to be Accepted, Pas, which results when using these skip lot procedures. For example, let's assume the lot size is 6,001 to 12,000 containers, and we use the single sampling plan of size 168 for normal inspection (Table I of § 42.109) along with an AQL for total defects of 6.5. The OC curve for this sampling plan is curve N and R3. Also, assume that the quality of the lot is 6.5 defects per hundred units. From curve N and R3, then, the Pa is determined to be 95 percent. Then refer to the graph in § 42.142 and locate Pa of 95 percent along the horizontal axis. From this point, proceed vertically to the curve and then horizontally to the left to the vertical axis. From this point on the vertical axis, Pas is determined to be approximately 981 percent for skip lot sampling and inspection. [41 FR 42641, Sept. 28, 1976. Redesignated at 42 FR 32514, June 27, 1977 and 45 FR 69424, Oct. 21, 1980 and further redesignated at 46 FR 63203, Dec. 31, 1981] 1 'This slight increase of Pas over Pa results because all the lots offered for inspection are no longer being inspected. Some lots are being "skipped", and thereby accepted, which would have been rejected if inspected. The largest increase in Pa which will occur using skip lot sampling is approximately 4 percent. It is for this reason that skip lot sampling procedures are applied only in instances where past production quality has been consistently high-as evidenced by the requirement to have 10 consecutively acceptable lots prior to a reduction in sampling rate-and where all requirements in § 42.120 also have been met. 13-006 0-83--49 § 42.142 Curve for obtaining Operating Characteristic (OC) curve information for skip lot sampling and inspection. 100 Percent of Lots Expected To Be Accepted When Using Skip Lot Sampling And Inspection, Pas NOTES: 1. This curve applies only to the specific skip lot sampling and inspection plan described in § 42.121 and § 42.123. 2. Pa and Pas are essentially equal for values under 50 percent. [41 FR 42641, Sept. 28, 1976. Redesignated at 42 FR 32514, June 27, 1977 and further redesignated at 45 FR 69424, Oct. 21, 1980 and 46 FR 63203, Dec. 31, 1981] § 42.143 Operating Characteristic (OC) curves for on-line sampling and inspection. (a) This section contains the Operating Characteristic (OC) curve for each of the one-line cumulative sum sampling bplans provided in Subpart D. The OC curve and the corresponding cumulative sum sampling plans are listed by AQL (b) Different plan parameters used to make acceptance and rejection decisions are provided for each AQL. (c) The curves show the ability of the various cumulative sum sampling plans to distinguish between good and bad portions of production. The interpretation of these curves for portions of production is similar to the interpretation of the OC curves for stationary lots as illustrated in § 42.140(c). |