Chapter 15 Basic Acceptance Sampling Procedure From : Book of Statistical Quality Control by D. C. Montgomery Prof . Ma.Teodora E.Gutierrez THE ACCEPTANCE-SAMPLING PROBLEM I ► lot-by-lot acceptance-sampling plans for attributes acceptance sampling: ► concerned with inspection and decision making regarding product THE ACCEPTANCE-SAMPLING(™K®Ø) PROBLEMII A typical application of acceptance sampling ► a shipment of product from a supplier ► Product: a component; raw material used in the company’s manufacturing process ► A sample taken from the lot: quality characteristic of the units in the sample is inspected ► A decisionis made regarding lot disposition o r c a l l e d lot sentencing ) ) accept: put into production reject: returned to the supplier or to some lot disposition action THE ACCEPTANCE-SAMPLING(™K®Ø) PROBLEM III Three aspects of sampling: 1.sentence lots,not to estimate the lot quality 2.simply accepts or rejects lots: Even if all lots are of the same quality, sampling will accept some lots and reject others 3.an audit( 100%) tool to ensure that the output of a process conforms to requirements THE ACCEPTANCE-SAMPLING(™K®Ø) PROBLEMIV Three approaches to lot sentencing: 1.accept with no inspection 2.100% inspection : ) inspect every item, removing all defective units ) Situation: extremely criticalcomponent 3.acceptance sampling THE ACCEPTANCE-SAMPLING(™K®Ø) PROBLEMV Acceptance sampling is useful in the following situations: ► Test is destructive(%û¾) ► The cost of 100% inspection is extremely high ► 100% inspection is not technologically feasible or ð ¾ ► many item to be inspected or the inspection error rate is high than 100% inspection ► the supplier has an excellent quality history ► potentially serious product liability(.k) risks ADVANTAGES AND DISADVANTAGES OF SAMPLINGI Advantage: comparing with 100%inspection ► less inspection ► reduced damage ► applicable to destructive testing ► fewer personnel are involved ► reduces the amount of inspection error ► the simple return of defectives often provides a stronger motivation to the supplier for quality improvements ADVANTAGES AND DISADVANTAGES OF SAMPLINGII Disadvantage: ► risks of accepting "bad" lots and rejecting "good" lots ► less information ► Acceptance sampling requires planning and documentation of the acceptance-sampling procedure TYPES OF SAMPLING PLANSI Variables: Chap. 16 Attributes: Chap. 15 1. single-sampling plan: (sample size, acceptance number)=(n, c) if #{defective}>c ⇒ reject thelot 2. double-sampling plan 3. multiple-sampling plan: the extension is sequential sampling LOT FORMATION ► Lots should be homogeneous: the same machine, the same operators, from common raw materials, at approximate the same time ► Larger lots are preferred over smaller ones ► Lots should be conformable(ʼ¾) tothe materials-handling systems used in both the supplier and consumer facilities('n) SINGLE-SAMPLING PLAN FOR ATTRIBUTESI ► N : size of a lot; (Ex: N = 10, 000) ► n: the sample size; (Ex: n = 89) ► c: the acceptance number; (Ex: c = 2) ► d: the number of observed defective items d ≤ c ⇒ acceptes the lot Operating characteristic(OC) curve SINGLE-SAMPLING PLAN FOR ATTRIBUTESII ► An importantmeasure of the performance of an acceptance-sampling plan ► display(•î) the discriminatory($ïIh¾) power of the sampling plan ► the probabilitythat a lot submitted with a certain fraction(½Iz) defective will be either accepted orrejected SINGLE-SAMPLING PLAN FOR ATTRIBUTESIII ► B(n, p) P{d defectives} = ► . Σ n d p (1 −p) d n− d Probability of acceptance: Pa= P{d ≤ c} = Σc . nΣ pd(1 − p) n−d d d=0 Ex: n = 89, c = 2, p = 0.01 ⇒ Pa = P{d ≤ 2} = 0.9397 n = 89, c = 2, p = 0.02 ⇒ Pa = P{d ≤ 2} = 0.74 100 lots: expect to accept 74 of the lots and reject 26 ofthem SINGLE-SAMPLING PLAN FOR ATTRIBUTESIV (a) ideal (b) sample size (c) c (a)discriminated perfectly between good and bad lots (b) c n=constant: the greater the slope of the OC curve, the greater is the discriminatory power (c)changing c does not dramatically change the slope of the OC curve DESIGNING A SINGLE-SAMPLING PLANI Type I and Type II errors: ► H0 : the lot quality is good Accept H0 Reject H0 ► State of the lot Good Quality of Lot Poor Quality of Lot (H0 is true) (H0 is false) Correct β=Type II error α=Type I error Correct To the design of an acceptance-sampling plan: require that the OC curve pass through two designated points DESIGNING A SINGLE-SAMPLING PLANII Σ . Σ c 1 −α = d=0 n d p (1 − p1) n−d d 1 (1 − α = the prob. of acceptance the lots with p1) Σc . nΣ β= pd(1 − p )2 n− d d 2 d=0 (β = the prob. of acceptance the lots with p2) DESIGNING A SINGLE-SAMPLING PLANIII ► 1 − α, β: two simultaneous equations are nonlinear ► there is no simple, direct solution ► (p1, 1 − α) = (0.01, 0.95), (p2, β) = (0.06, 0.10) ► Any two points on the OC curve could be used to define the sampling plan ► p1 =AQL (producer’s risk), p2 =LTPD (consumer’s risk) ► AQL: acceptable quality level ► LTPD: lot tolerance percent defective = RQL(rejectable quality level)=LQL (limiting quality level) Designing a Single-Sampling PlanIV DESIGNING A SINGLE-SAMPLING PLAN ► ► AQL: acceptable quality level the poorest level of quality for the supplier process LTPD: lot tolerance percent defective the poorest level of quality that the consumer is willing to accept in an individual lot Designing a Single-Sampling PlanVI RECTIFYING INSPECTION I ► Rectifying inspection: Average outgoing quality(AOQ)(): the evaluation of a rectifying sampling plan AOQ RECTIFYING INSPECTION(ƑFL™)II AOQ = Pap(N −n) N n/N →0 c Pap ► n: after inspection, contain no defectives ► N − n: contain no defectives if the lot is rejected ► N − n: contain p(N − n) defectives if the lot is accepted AOQ: Rectifying Inspection(ƒfl™)III DOUBLE-SAMPLING PLANSI DSP ► DSP: a procedure; a second sample is required before the lot can be sentenced. ► four parameters: n1 =sample size on the first sample c1 =acceptance number of the first sample n2 =sample size on the second sample c2 =acceptance number of the second sample ► di =the number of defectives in the ith sample Double-Sampling PlansII DOUBLE-SAMPLING PLANSIII OC curve: Pa = PI a+ PII a Ex: (n1, n2, c1, c2, p) = (50, 100, 1, 3, 0.05) ► PaI= Σ1 d1=0 . Σ n (0.05)d1(0.95) d1 PaII = P{d =1 2, d ≤ 21 50− d1 = 0.278 }+ P{d =13, d = 0} 2 = P{d1 = 2} · P{d2 ≤ 1} + P{d1 = 3} · P{d2 = 0} =0.0107 Pa = PI a+ PII = a 0.2897 DOUBLE-SAMPLING PLANSIV [PaI(N − n ) 1+ P (NaII− n − n )]p 1 2 N II ATI = n P1 +aI (n + n 1 )P 2 a + N(1 −P ) a AOQ =