For attributes acceptance sampling, you can use OC curves to explore how sampling plans and assumed product quality impact the probability of accepting a lot. Use the text boxes, sliders, or fraction defective handle on the OC curve to set and adjust your sampling plan.
The following options are available for single sampling OC curves:
Sampling Type
Enables you to select Lot Sampling or Binomial Sampling.
Lot Sampling
Enables you to specify and explore an acceptance plan based on a fixed lot size.
Binomial Sampling
Enables you to specify and explore an acceptance plan for a continuous process or other situation where the binomial distribution is appropriate.
Lot size (N)
(Available only for Lot Sampling.) Specifies the size of the lot that you are sampling from.
Sample Size (n)
Specifies the number of units for inspection.
Acceptable failures (c1)
specifies the number of allowable failures. If the number of observed defects is greater than c1, then the lot is rejected.
Acceptable Quality Limit (AQL)
Specifies the poorest quality limit that is acceptable. Enter the highest proportion of defects, on average, that is acceptable. One desires a sampling plan with a high probability of accepting a lot with a defect rate less than or equal to the AQL.
Rejectable Quality Limit (RQL)
Specifies the poorest quality for an individual lot. Enter the highest proportion of defects that is acceptable for an individual lot. RQL must be greater than or equal to AQL. RQL is also known as the limiting quality level (LQL) or the lot tolerance percent defective (LTPD). The consumer desires a sampling plan with a low probability of accepting a lot with a defect rate greater than or equal to the RQL.
Fraction Defective
Specifies the expected fraction defective in the lot.
Probability of Acceptance
Specifies the probability of accepting the lot given the sampling plan as defined. The Probability of Acceptance updates as the specifications are changed, but you can also adjust this value directly. When you adjust the Probability of Acceptance directly, the Fraction Defective value is updated.
Alpha
Specifies the Type I error. This is the probability of rejecting a lot that has a defect level equal to the AQL. This is also known as the producer’s risk.
Beta
Specifies the Type II error. This is the probability of accepting a lot that has a defect level equal to the RQL. This is also known as the consumer’s risk.
The following options are available for double sampling OC curves:
First Sample
Contains the following settings for the first sample:
Number inspected (n1)
Specifies the number of units inspected in the first sample.
Acceptable failures (c1)
specifies the number of allowable failures in the first sample.
Second Sample
Contains the following settings for the second sample:
Number inspected (n2)
Specifies the number of units inspected in the second sample.
Acceptable failures (c1+c2)
specifies the total number of allowable failures.
Acceptable Quality Limit (AQL)
Specifies the poorest quality limit that is acceptable. Enter the highest proportion of defects that is acceptable.
Rejectable Quality Limit (RQL)
Specifies the poorest quality for an individual lot. Enter the highest proportion of defects that is acceptable for an individual lot. RQL must be less than or equal to AQL.
Fraction Defective
Specifies the expected fraction defective in the lot.
Probability of Acceptance
Specifies the probability of accepting the lot given the sampling plan as defined. The Probability of Acceptance updates as the specifications are changed, but you can also adjust this value directly. When you adjust the Probability of Acceptance directly, the Fraction Defective value is updated.
Alpha
Specifies the Type I error. This is the probability of rejecting a lot that has a defect level equal to the AQL
Beta
Specifies the Type II error. This is the probability of accepting a lot that has a defect level equal to the RQL.