Acceptance Sampling [Weightage of Marks = 06, Teaching Hours = 04 7.1 Acceptance Sampling: Concept, Comparison with 100% inspection, Operating Characteristics Curve (O.C. Curve) 7.2. Different Types of Sampling Plans, Sampling Methods, ‘About this Chapter At the end of this chapter, students will be able to: * Understand acceptance sampling or samplin; limitations. + Draw O.C. curve for given values of parameters and s Merits and Demerits of Acceptance Sampling. 1g inspection with its importance, advantages and ampling inspection. 5 1 ACCEPTANCE SAMPLING QUESTION | 4. why 100% inspection is generally not preferred in the Industry for mass production? y manufactured component or part, whether to be accepted or y gut at many stages during the entire manufacturing process. The f raw material, semifinished part or finished part etc. * To take decision about an rejected, inspection is carrie part inspected may be in the form o wo ways to carry out inspection activity: «Following are the t 1. 100% inspection. 2. Acceptance sampling 0 + In '100% inspection’, all the parts sampling or Sampling Inspection’, on! the entire lot of parts- + acceptance sampling or Sampling inspection is defined as, "the process of inspecting or ae sneting 2 sample containing @ small Quannty of products chosen from entire lot of products gp TT dl tl the basis of number of defective parts found - Sampling inspection. re subjected to inspection, whereas, in ‘Acceptance \y a ‘sample’ of few parts is drawn for inspection from in that sample.” + ifthe number of defective parts does not exceed the predefined acceptance number level, the said lot is accepted, otherwise rejected. + In be cceptance eres Is mor dared tems)pars/producta nes “> Convo! Me [7.44 Advantages and Disadvantages of Acceptance Sampling as ceva to 100% Inspection OR Comparison between | ¢ Sampling and 100% Inspection QUESTION ih 100% inspection may {je beIndustrial Engineering and Quallty Contro! (A) Advantages of Acceptance Sampling as Comp: ared to 100% Ins! ‘Acceptance Sampling pection: 100% Inspection sr.| Comparative Point | Acceptance Sampling or os No. Sampling Inspection | 1. | cost of inspection _| Very less. _| High. _— | 2. Time required to Less. High. : | carry out inspection _| i [oT 3. _ Inspection fatigue No. —lyess a 4, Inspection error It can be minimised to great | Not possible to minimise. problem extent. feels s |eee ros) e 5. | Scheduling and As only few items undergo | As all “parts (100%) 7 oc dispatch dates inspection, the accepted lot | inspected, this method takes can be sent to market in least} more time = to_— sort oul time. It improves the | non-defective products, which scheduling and dispatch | can be sent to market. So, dates. scheduling and dispatch dates cannot be improved. Effective pressure on | More. Less. quality improvement (B) Disadvantages of Acceptance Sampling as as Compared to 100% Inspection: 1 _ State limitations of acceptance sampling, "QUESTION Comparative Point Sr. Acceptance Sampling or 100% Inspection No. Sampling Inspection 1. | Loss of producer, A lot containing less defectives may |No loss except damage while rejecting a lot.. | be rejected due to more defectives | during transportation. found in the sample drawn from the 7 lot. Producer will suffer the loss. 2. Loss of consumer, | A lot containing more defectives | No loss of consumer, as all | while accepting a may be accepted due to less | parts available in market lot. | Product. number of defectives found in the sample drawn from the lot. Thus, the entire lot containing more number of defectives will be sent to market for sales. Then, many consumers will suffer due to defective articles or products received on purchase. Also, due to consumer complaints regarding replacement or repair of faulty products, the producer has to suffer loss, because of cost involved in transportation for replacing defective products and cost involved in repairing defective Product at consumer's place. For retaining profit, the producer will Pass on these additional expenses on consumer by raising the price of are non-defectives, because of 100% inspection.ii ‘Acceptance Sampling ‘Industrial Engineering and Qualty Control 1 — g| Precautions to be Taken for Random Sampling) is to be made on the basis of while carrying out random i Since a judgment about either accepting or rejecting ied :. sample drawn from a lot, following precautions must be : ‘sampling. ining the 1. Sample should be collected from all sides and different depths of container containing lot. It of 2. Sample size (n) should be sufficiently large to obtain most accurate and reliable resu! sampling inspection. TYPES OF SAMPLING PLANS a 7 "QUESTIONS ; $14) 1 2 Enlist the types of sampling plans. State different types of sampling plans and explain any one of them. * Various types of Sampling plans are, 1. Single sampling plan, 2. Double sampling plan, and Multiple sampling plan. 7.4 SINGLE SAMPLING PLAN : + When a decision of acceptance or rejection of lot is made on the basis of only one sat it is called as single sampling plan, To understand the stepwise procedure of carrying out single sampling plan, refer the flowchart as shown in Fig. 7.1. Let us consider following three parameters for single sampling plan. 1. N= Lotsize, 2. n = Sample size, 3. C ‘Acceptance number = Maximum number of allowable defectives in the sample of size 'n Inspect a sample of i pieces 7-4: Flowchart or Block diagram illustrating Single sampling plan ‘one sample. ° M f i oe of carrying out single sampling pian js illustrated with the help of an example given + Consider the following example illustratin, ig single sampling plan. Let, N= Lot size or Number of items in the lot = S00. n= Sample size or Number of items in the sample = 35 C = Acceptance number for the sample = Maximum number of d items in the sample, that will permit the acceptance of lot 2 etyindustrial Engineering and Quality Controt nipol__ A tance Samptin procedurt 1. Take or draw a > Ifthe sample eae eeerene from a lot of 500 items and inspect it. Ifthe NM ive items less than 2, then accept the lot. 3. sample contains defective items more thi ~ an 2, then reject the lot. 7.5 DOUBLE SAMPLING PLAN : QUESTIONS 4. Explain double sampling plan with suitable example. 2. Explain the double sampling plan with a suitable block diagram. 3. Explain double sampling plan, When a decision of acceptance or rejection of lot is made on the basis of two samples combined, it is called as double sampling plan. «Alot may be accepted at once, if the first sample is good enough and rejected at once, if the sample is bad enough. + If the first sample is neither g' reject the entire lot, then a secon: made on the basis of first and seco * Note that, in case of double sampling plan, the products in second sample), is always taken as twi products in first sample). Therefore, M2 = 2 M1 + Refer the flowchart shown in Fig. 7.2 to understand ster double sampling plan. entire lot nor bad enough to jood enough to accept the us fhe same lot and decision is d sample is drawn from tl nd sample combined. e size of second sample (i.e. ice the size of first sample (i. Number of . Number of pwise procedure of carrying out + Parameters: - Let, n; = Number of pieces in the first sample. Ci = Acceptance number for the first sample = Maximum number of defectives, that wai permit the acceptance of [ot on the basis of first sample. ‘ces in the second sample. nz = Number of pies m+n, = Number of pieces 1 the two samples combined. | = Acceptance numbar for the two samples combined = Maximum number of 2 = Befectives, that will permit the acceptance of lot on the basis of first and second Sample combined. (& Trepecta sample of ‘nj pieces ‘number of defectives in the first and second ‘samples combined : [Foon lustrating Double sampling plan Fig. 7.2: Flowchart or Block diagram78 REESAMPLINGPLANS = ‘single | Double | Multiple Som? | ‘Sampling Plan Plan A Industrial Engineering and Quality Control 7.8 COMPARISON BETWEEN TH Comparative Point ‘sampling Plan pg Pei age ec Less acceptable, | Accer as it gives Most acceptable i | a Cay one ‘ance | (Gives 11? | indecision continues, ee i chance). for a long perio i; teeter | more number of | Siete oes : | samples are drawn for te Oeeced | making decision. | i ne one ae In between. Lowest. | | © @. | Numb Multiple (i.e. Three or One. Two. * wn aes more than three). jr 5. Variability of inspection | Constant. Variable. Highly variable. | (load Number of items | | [Le. number of items undergoing | | inspection ci inspection} j= jergoing insp 3] depends upon behaviour of first | sample. | 6. | Estimation about quality | Best. Intermediate. _| Difficult to estimate. ic a a = 7. | Decision of ‘Acceptance’ | One sample Two samples —_| Multiple samples or ‘Rejection’ is based combined. combined. on 1 | - 7.9 CHARACTERISTICS OF GOOD SAMPLING PLAN 5 ‘+ [Link] sampling plan should have following characteristics/properties. Simple to operate, Economy of inspection i.e. economical inspection. 1 2 3. Flexible enough to reflect changes in size and quality of product submitted. 4. Plan should protect both ‘producer and consumer’: : E Less difficult in trainin i ° — the prescribed or sampling plan’. inspectors about ‘how to use # cleat 7-40 FACTORS TO BE CONSIDERED, WHILE SELECTING A SAMPLING PLaN Protection to both, ‘consumer and producer! 2. Cost of inspection, Availability of ‘inspectors and facilities’ : - correct To train inspectors about how to use diferent sampling plans“ correc Need for quick and reliable estimates. Administrative convenience.industrial Engineering and Quality Control 79 oe : Acceptance 741 0.C. CURVE Tx «own QUESTIONS : iat is meant by O.C. curve? Draw ideal and actual O.C. curve and Explain. | (i) Producer's tisk, (ii) Consumer risk. 2. Explain Ideal and Actual .C. curve with all significant points and regions. eH) 3. Draw a neat sketch of O.C. curve. Show the different regions and explain the meaning of these 18) regions. 4. Sketch ideal and actual O.C. curve showing all parameter: __Explain 0.C. curve. Draw ideal and actual 0.C. curves.__— 5 on it and define them. About Operating Characteristic curve (0-C- curve): urve (0.C. curve) for an attri st the probability of acceptance. curve shows the probability Ps, such © Operating Characteristic ¢\ bute sampling plan is a graph of fraction defectives in a lot again = For fraction defective p' in a submitted lot, the 0.C. that, the lot will be accepted by sampling plan. Importance of 0.C. curve: 1. 0.C. curve of an acceptable sampling plan show: between good lots and bad lots. \dge and com uality level of submitted product. 5 the ability of plan to di inguish 2. 0.C. curve is used to ju pare the performance of various sampling plans over a range of possible 4 Shape of O.C. curve: «Shape of O.C. curve dep’ (a) N= Lot size, from whicl ends upon following parameters. ‘h, samples are drawn. (b) n= Sample size. ) Acceptance number. + By changing the parameters +N’, ‘n’ and ‘C’, different sampling plans, the shape of 0.€. curve will differ. curve: different sampling plans are obtained and for Construction of 0.C. + To construct an 0.C. curve, lots with change in quantity 0! Poisson’s distribution. + Acceptance plan is desi through two stipulated P' we should know the mathematical probability of accepting the f percent defectives. This can be obtained from the table of igned in such a way that, Its corresponding 0.C. curve will pass ints agreed upon by both, consumer and producer. Ideal 0.C. curv! * Inthe Fig. 7-3, AP! * We observe that, if propor of lot will be 95% i.e. Po * But, if proportion of defectives Is Mo! 7 . jeal, if all ood Ik Oc. curve lean ove (geal 0. eee accepted and bad lots are rejected. But, curve is not practic BCD shows ‘Ideal 0.C. curve’ rtion of defectives is less than AQL, 1a 8. QL, then probability of acceptance re than LTPD value, then Pa = 0. * Therefore, Ideal 0.C- possible.‘Acceptance Sampling 31 Engineering and Quality Controt 7.10 Actual 0.C. curve: + Practically possible 0.C. curve is called as actual 0.C. curve: + In fig. 7.3, A'PCB" shows ‘Actual 0.C. curve’. + The degree, to which, an actual 0.C. curve approximates the ideal upon sample size (n) and acceptance number (C). 10.C. curve, depends Note: The meanings of different points inthe curve are given in table below. 1 Indiference. ‘quality region 1 Objectional (Iau) igus region Consumer's tisk AQL LTPD) 10 Process average or Proportion of defectives —» Fig. 7.3: 0.C. curve Meanings of various points plotted in the figure 7.. sr. | Point Shown | Probability of | "Description | No. | in Figure | jet | 100% Zero defectives. 7 2| 95% 5% defectives. 7 © 8 [2 100% defectives. 2 7 4 8 10nmemnn codename Ll ___| Indicates "Acceptable quality level" | 6 -_ 50% | Indicates “indifference quality level". SCS [> LTPD Lower Indicates "Objectionable OR Rejectionable qualitY bac} Proamare me 1 risk rote na ute @ good lot. Therefore | 9. | Consumer's = Indicates risk of accepting a bad lot we risk | | consumer may suffer. 7 + Accordingly, an 0.C. curve can be divided into three regions. (a) Acceptable quality region. (b) Indifference quality region (c) Objectionable quality region.Indust Acceptance Sampling 32 Industrial Engineering and Quality Controt 7 e efec ‘ems found in 3 rejected lots = 16 (say) Let, total number of defective item ut opted and rejected lots = 20 + 16 = 36. This © © Total number of defective tems | _ value represents p’ on to consumer, even though, they have 20 defective items. | tots sed 4 ne acer acts bre made to undergo 100% inspection and all the defective items (16) cted 3 lots ir ” ound la Pe rccreces aie gre replaced by non-defective items. Therefore, total number Of defective ‘tems in all 10 lots (i.e. 7 accepted and 3 rejected lots) will be, 20 + 0 = 20 | defective items. This value of defective items (i.e. 20) represents AQQ, which is less than p’ (i.e. 36). 7 = ——— . 5. AOQL: . . AOQL is ‘Average Outgoing Quality Level’, Fovany sample size and acceptance number, there is @ maximum value of AOQ beyond twhich, the average fraction defectives will not rise, no matter, how bad is the quality of lots, tunen they undergo sampling inspection to sort out acceptable lots and rejected lots, which is followed by 100% inspection of rejected lots to replace all defectives found in rejected lots by non-defectives. Therefore, AOQL can be referred as, ‘maximum possible value of average percent defectives in the outgoing products after inspection and rectification’. ~ Tt means that, if the incoming quality is bad, the sampling plan will cause all the lots to be rejected, 100% inspected and rectified. Therefore, the outgoing quality will be good. Therefore, even if the incoming quality is very good or very bad, the outgoing quality will tend to be very good. Between these two lots, there will be a point, at which, the percent of defectives will its maximum. This point is known as average outgoing quality level (AOQL). 7.11.1 Difforen: tween AQL and AOQL| AQL quality level. 2. It is defined as ‘the maximum percent | 2. It is referred as ‘maximum possible value 7 aogt Tt is average outgoing quality level. 1 “It is accept: defectives, which can be considered as| of average percent defectives in the) satisfactory for sampling inspection. outgoing products after inspection and 3. It hi — rectification’. | : i hes high probability of acceptance up to | 3. It has less probability of acceptance. | . It is the maximum value of AOQ, veya Which, the average percent defectives | cannot increase in the outgoing products, after inspection of all lots, followed by | = — |__ rectification of rejected lots. _ = 1.2 Characteristics of 0,C. Curve| 1. O.€. ty of sar lots. 2. " eoeaat ones of ‘acceptance number’ (C) and ‘sample size’ (mn), 0.C. curves sample sen (7) eee niet ot lot size {H) do not differ considerably. 1 Indicates that, 3. For large Sane much more important as compared to lot size (N). bad lots, if loping: O.C. curve will give better discrimination between good lots and q Conditio OU prada il le il) Cx a on sro, NecePtance number is directly proportional to sample size -C. curve will be more steeper for large values of sample size (n) and acceptance number (C). Steeper means slo ; 7 ye of lit aking an angle of 90° with the horizon oP ine, almost approaching vertical line, making 9 4. It is the maximum proportion of defectives, | 4. which the consumer finds definitely acceptable. curve represents the ability of sampling plan to distinguish between good lots and bad_ Industrial Engineering and Quality Control m3 —— 7 = hee Sampling .3_ Significance of 0.C. Curve fz 1. 0.C. curve of an acceptable sampling plan shows the ability of plan to distinguish between good lots and bad lots. ” sneuish 2. 0.C. curve is used to judge and compare the performance of various sampling plans over a range of possible quality levels of submitted products. 3. It provides a method for evaluating sampling plans. 4. It finds out the changing quality and condition of incoming materials. 5. It explains the risks, present compulsorily in a sampling plan at each level of quality. No sampling plan can be free from these inherent risks. . 7.12 PRODUCER'S RISK 2 (s-14, 15) QUESTION 1. Explain Producer's risk. + Producer's risk can be defined as, “the probability of rejecting a good lot, which otherwise, + would have been selected (accepted). + It means that, lot will be rejected, if the proportion of defective products found in a sample drawn from corresponding lot is more than acceptance number or set value, even though, the maximum number of products of lot falls within an acceptance allowable range. The designated set value is generally AQL. + Producer's risk can also be defined as, “the probability that, a lot having AQL will be rejected”. + This risk is generally taken as 5%, such that, Probability of acceptance, P, = 1 ~ 0.05 = 0.95. + Even if quality is good, some lots are rejected due to results of sampling plan, then producer has to suffer. + Therefore, producers should be protected against rejection of relatively better products. + Producers can decrease risk by producing product at better quality level than specified AQL without much increase in production cost. 2.1 Example of Producer’s Risk| * In producer's risk, a lot is rejected, even though, the lot or number of products in that lot falls within an acceptable allowable range. * For example: Manufacturer of pens rejects a lot of high quality pens due to more number of defective pens found in the sample drawn from the batch. 7.13 CONSUMER'S RISK ; Sp QUESTION 1._Explain Consumer's risk. __ 7 * Consumer's risk is defined as, “the probability of accepting the defective lots, which otherwise would have been rejected”. + Even if quality is bad, some lots are accepted due to results of sampling inspection, then consumer has to suffer. + Saying that, Ps, , = 2.5 means that, the consumer does not want a worse quality containing more than 2.5% defectives and he would, at the most accept, 10% of lots containing defectives less than 2.5%. * Here, 10% of lots accepted Is calculated by value 0.1 given in the above relation of Probability of acceptance (i.€. Pay, = 2-5)- Multiply the given value 0.1 by 100 to obtain percentage of lots ___accepted, i.e. 0.1 x 100% = 10%. 7.13.4 Example of Consumer's Risk| Tm consumer's risk, a lot is accepted, even though, the lot or number of products in that lot does not fall within an acceptable allowable range. For example: A purchased house of high quality indicates that, even though all the essential accessories for house are much durable, but plumbing work is failed, then it becomes ‘nsumer's risk.