880 likes | 1.49k Views
Quality Control. Chapter 10- Acceptance Sampling Systems. PowerPoint presentation to accompany Besterfield Quality Control, 8e PowerPoints created by Rosida Coowar. Outline. Lot-by-lot acceptance sampling plans for attributes Acceptance sampling plans for continuous production
E N D
Quality Control Chapter 10- Acceptance Sampling Systems PowerPoint presentation to accompany Besterfield Quality Control, 8e PowerPoints created by Rosida Coowar
Outline Lot-by-lot acceptance sampling plans for attributes Acceptance sampling plans for continuous production Acceptance sampling plans for variables
Learning Objectives When you have completed this chapter you should be able to: Determine the sampling plan using ANSI/ASQ Z1.4. Know the switching rules for ANSI/ASQ Z1.4. Categorize the various sampling plan systems in terms of lot-by-lot, continuous production, attributes or variables.
Learning Objectives-cont’d. When you have completed this chapter you should be able to: Describe the various sampling plan systems and know their function (advantages, disadvantages, purpose etc.). Determine the sampling plan using the Dodge Romig Tables Construct the OC Curve for a chain sampling plan.
Learning Objectives-cont’d. When you have completed this chapter you should be able to: Be able to use the Shainin Lot Plot Method. Determine the sampling plan ANSI/ASQ S1 Determine whether a lot is accepted or rejected using ANSI/ASQ Z1.9
Lot-by-Lot Acceptance Sampling Plans for Attributes • Devised in 1942 at Bell Labs • Later became MIL-STD-105E • ISO-2859 • ANSI/ASQ Z1.4 • Wording & terminology changes • Additional tables added
ANSI/ASQ Z1.4 • This standard is applicable to: • End items • Components and raw materials • Operations • Materials in process • Supplies in storage • Maintenance operations • Data or records
ANSI/ASQ Z1.4 • Intended to be used for a continuing series of lots, but may be designed for isolated lots. • Standard provides for single, double, and multiple sampling plans. • Provision is also provided for normal, tightened, or reduced inspection. • Plan is specified by the AQL, and sample-size code.
ANSI/ASQ Z1.4 • Normal inspection is used at the start of inspection with changes being a function of recent quality history. • Tightened inspection: • Generally used when producer’s recent quality history has deteriorated.
ANSI/ASQ Z1.4 • Reduced inspection: • Used when the producer’s recent quality history has been exceptionally good. • Decision concerning what type of plan to use is left to the responsible authority. • Nonconformities are classified such as critical, major…
ANSI/ASQ Z1.4 • AQL • The AQL is the most important part of the standard since the AQL and sample size dictate the plan used. • AQL is defined as the maximum percent nonconforming that can be considered satisfactory as a process average. • Satisfactory = Producer’s risk, α • Usually 0.05
ANSI/ASQ Z1.4 • AQL • AQL is specified by contract or responsible authority. • AQL may be determine from historical data, empirical judgment, engineering information, experimentation, producer’s capability, consumer’s requirements.
ANSI/ASQ Z1.4 • AQL • AQL is specified by contract or responsible authority. • AQL may be determine from historical data, empirical judgment, engineering information, experimentation, producer’s capability, consumer’s requirements. 17
ANSI/ASQ Z1.4 • Sample Size • Determined by lot size and inspection level • Use of sample-size code • Inspection level is determined by the responsible authority. 17
ANSI/ASQ Z1.4 • Sample Size • Different levels of inspection provide approximately the same protection to the producer, but different protections to the consumer. • Plan provides for “special” levels to be used where relatively small sample sizes are necessary and large sampling risks must be tolerated. (S-1, S-2, S-3, S-4) 18
ANSI/ASQ Z1.4 • Implementation • Determine lot size • Determine inspection level • Find sample-size code letter in table • Determine AQL 19
ANSI/ASQ Z1.4 • Implementation cont’d. • Determine type of sampling plan • Find sampling plan in appropriate table • Start with normal inspection and change to tightened or reduced based on switching rules 20
AQL What an AQL means is that as long as a supplier maintains his/her process average (% defective) at the assigned AQL or lower, there is a very high probability that shipments from that supplier, when inspected using ASQ Z1.4 sampling plans, will be accepted. 21
AQL By the same token, there is a very high probability that shipments, when inspected using ASQ Z1.4 sampling plans, will be rejected if a supplier's process average (% defective) remains higher than the assigned AQL. 22
ANSI/ASQ Z1.4 • Single sampling plans • Start by identification of AQL, lot size, inspection level, and type of sampling plan. • Locate appropriate plan parameters in tables. • Double and Multiple Sampling • Similar procedures as single sampling 23
ANSI/ASQ Z1.4 • Normal, Tightened, and Reduced Inspection • Start with normal inspection • Switching Procedures • Normal to tightened • Institute when 2 out of 5 consecutive lots not accepted on original inspection • Tightened to normal • Institute when 5 consecutive lots accepted • If not, then discontinue inspection under this plan 24
ANSI/ASQ Z1.4 • Switching Procedures • Normal to reduced • Institute when: (all of the following) • Preceding 10 lots on normal inspection have been accepted • Total nonconforming in preceding 10 samples is less than number identified in Table 10-5. • Production is at a steady rate • Reduced inspection is considered desirable by responsible authority 25
ANSI/ASQ Z1.4 • Switching Procedures • Reduced to normal • Institute when: (any of the following) • A lot is not accepted • Sampling procedure terminates with neither acceptance or rejection criteria have been met • Production is irregular or delayed • Other conditions 26
ANSI/ASQ Z1.4 • Standard is designed for use where units of product are produced in a continuing series of lots or batches. • If a sampling plan is desirable for a lot or batch of an isolated nature, it should be chosen based on the Limiting Quality (LQ) and consumer’s risk, b. • These tables are included in the standard, but not the textbook. • Or use ANSI/ASQ Standard Q3 27
ANSI/ASQ Standard Q3 • Standard is used for inspection of isolated lots by attributes. • Provides indexed tables by Limiting Quality, LQ. 28
ANSI/ASQ Standard Q3 • Two Schemes: 1. Used for lots that are isolated or mixed or that have an unknown history as far as all parties know. • Lot size and LQ must be known. • Nominal values of the LQ are based on b=0.10 29
ANSI/ASQ Standard Q3 • Two Schemes: 2. Used when a vendor is producing a continuous stream of lots and sends one or a few to a customer who will consider them as isolated lots. • Occurs frequently when purchasing small quantities of raw materials. 30
Dodge-Romig Tables • A set of inspection tables for lot-by-lot acceptance sampling by attributes. • Tables are based on Limiting Quality (LQ) and Average Outgoing Quality Limit (AOQL). • Tables for single and double sampling • Advantage that the plan provides a minimum amount of inspection for a given inspection procedure. • Desirable for in-house inspection 31
Dodge-Romig Tables • Limiting Quality – LQ • Based on the probability that a particular lot, which has percent nonconforming equal to the LQ, will be accepted. • Probability is the consumer’s risk, β = 0.10 • LQ plans give assurance that individual lots of poor quality will rarely be accepted. 32
Dodge-Romig Tables • Observations about the LQ tables: • As the lot size increases, the relative sample size decreases. • The tables extend until the process average is one-half of the LQ. • As the process average increases, a corresponding increase occurs in the amount inspected. 33
Dodge-Romig Tables • Average Outgoing Quality Limit – AOQL • Applicable when the inspected lot is a convenient subdivision of a flow of product for materials handling purposes (non-homogeneous) • AOQL plan limits the amount of poor outgoing quality on an average basis, but gives no assurance on individual lots. 34
Dodge-Romig Tables • Process average • Obtained by the same techniques as for p-charts • First 25 lots the average percent nonconforming is obtained • Type of nonconformance • Dodge-Romig tables do not provide for different categories or tightened inspection • Different LQ or AOQL values may be used 35
Chain Sampling Inspection Plan • A special type of lot-by-lot acceptance sampling plan for attributes designated as “Chain Sampling Plan ChSP-1”. • Applicable to quality characteristics which involve destructive or costly tests. • Use of small sample sizes • Acceptance number, c=0 • Poor shape of OC curve at the producer’s risk 36
Chain Sampling Inspection Plan • Chain sampling uses the cumulative results of several samples • Determination of the value of i, the number of previous samples, is determined by analysis of the OC curves for a given sample size. 37
Chain Sampling Inspection Plan • Procedure: • For each lot, select a sample of size n and test each for conformance to specifications. 38
Chain Sampling Inspection Plan • Procedure: • If the sample has 0 nonconforming units, accept the lot; if the sample has 2 or more nonconforming units, do not accept the lot; and if the sample has 1 nonconforming unit, it may be accepted provided that there are 0 nonconforming units in the previous isamples of size n. 39
Chain Sampling Diagram Current Lot i previous lots n=5 c=0 LastLot Second- to-Last Lot Accept if 0 Nonconforming Units in n =5 or if 1 Nonconforming Unit Zero Nonconforming Units in i Previous Lots & Figure 10-4 Chain Sampling Diagram 40
Chain Sampling Inspection Plan Conditions for use of the Chain Sampling Plan: • The lot should be one of continuing series of product that is sampled in substantially the order of production • The consumer can normally expect the lots to be essentially the same quality. 41
Chain Sampling Inspection Plan Conditions for use of the Chain Sampling Plan: • The consumer has confidence in the producer not to occasionally send an unacceptable lot that would have the optimum chance of acceptance. • The quality characteristic is one that involves destructive or costly tests. 42
Sequential Sampling • Used for costly or destructive tests • Subgroup size of 1 – item by item plan • Sequential sampling is similar to multiple sampling, except sequential sampling can, in theory, continue indefinitely. • In practice, the plan is truncated after the number inspected is equal to three times the number inspected by a corresponding single sampling plan. • Based on the Sequential Probability Ratio Test 43
Skip Lot Sampling • Designated SkSP-1 and based on AOQL. • A single sampling plan for minimizing inspection costs when there is a continuing supply of lots of raw materials, component parts, subassemblies, and finished parts from the same source. • Applicable to chemical and physical characteristics that require laboratory analysis. 46
Procedure for SkSP-1 Plans Begin by inspecting every lot When i consecutive lots are found to be conforming Discontinue inspection of every lot Inspect a fraction, f, of the lots In a random manner When an inspected lot is found nonconforming Figure 10-7 Procedure SkSP-1 plans 47
ANSI/ASQ S1 • Provides procedures to reduce the inspection effort when the supplier’s quality is superior. • A skip-lot scheme used in conjunction with attribute lot-by-lot plans given in ASQ Z1.4 48
ANSI/ASQ S1 • Requirements to use ANSI/ASQ S1 • Have a documented system for controlling product quality and design changes • Have instituted a system that is capable of detecting and correcting changes that might adversely affect quality. • Not have experienced an organizational change that might adversely affect quality 49
ANSI/ASQ S1 • Additional requirements of the product: • Be of stable design • Have been manufactured on a continuous basis for at least 6 months unless agreed to a longer period. • Have been on normal and reduced inspection during the qualification period. • Have maintained a quality level at or less than the AQL for at least 6 months. 50