Use of Historical Quality Data to determine Sample Size

P

pangchiaboon

#1
Hi all, i find difficult to understand the standard sampling plan in detail, may be is because i do not have any statistic background. To make thing simple, i plan to base on historical quality data to determine the sample size.

For example we buy carton from supplier for medical device packaging, base on Jun 09 - Jun 10 data, let say the total carton loaded to production was 10,000pcs, total rejected carton segregated from production was 50pcs, means failure 0.5%. Can we treat this 0.5% as the smallest sample size that we should sample from an incoming lot in next receiving from supplier? or may be for better assurance we high up the % to 0.6% for better assurance.

If today we received a new lot from supplier with Qty 5000, sample size (0.6%) = 30pcs and C=0. Is is a acceptable method?

To be more meaningful, the supplier quality performance data will have to be reviewed by quaterly basis to minitor the supplier quality performance, at the same time adjust incoming sampling plan to change the sampling sensitivity. Base on quarterly analysis, we can also identify what are the major supplier rejection symptoms that we should focus in receiving sampling inspection. I believe we can use this method to justify why we use a specific sample size, don't you think this is more practical instead of apply complicated sampling plan like 1.9, 1.4, C=0 ?

Please advise. Thanks.
 
Elsmar Forum Sponsor

Stijloor

Staff member
Super Moderator
#2
Hi all, i find difficult to understand the standard sampling plan in detail, may be is because i do not have any statistic background. To make thing simple, i plan to base on historical quality data to determine the sample size.

For example we buy carton from supplier for medical device packaging, base on Jun 09 - Jun 10 data, let say the total carton loaded to production was 10,000pcs, total rejected carton segregated from production was 50pcs, means failure 0.5%. Can we treat this 0.5% as the smallest sample size that we should sample from an incoming lot in next receiving from supplier? or may be for better assurance we high up the % to 0.6% for better assurance.

If today we received a new lot from supplier with Qty 5000, sample size (0.6%) = 30pcs and C=0. Is is a acceptable method?

To be more meaningful, the supplier quality performance data will have to be reviewed by quaterly basis to minitor the supplier quality performance, at the same time adjust incoming sampling plan to change the sampling sensitivity. Base on quarterly analysis, we can also identify what are the major supplier rejection symptoms that we should focus in receiving sampling inspection. I believe we can use this method to justify why we use a specific sample size, don't you think this is more practical instead of apply complicated sampling plan like 1.9, 1.4, C=0 ?

Please advise. Thanks.
Can someone help?

Thanks!

Stijloor.
 

Statistical Steven

Statistician
Staff member
Super Moderator
#3
The defect rate does not equal the sample size. If we can assume (safe assumption) that we can tolerate a 0.5% defect rate on average (this is called the AQL), we can develop a sampling plan. Depending upon the maximum percent defective (LTPD) that is acceptable, the sample size is calculated. The following are some examples for c=0

N AQL LTPD
10 0.51% 20.5%
20 0.25% 10.9%
30 0.17% 7.4%
50 0.10% 4.5%
100 0.05% 2.3%

The sample size is independent of the lot size (in general). You can use a standard like Z1.4 that gives you sample sizes that are lot size dependent.
 
P

pangchiaboon

#4
Thanks Steven.
One thing that confusing me is why we still use AQL for C=0 sampling plan. AQL is the max rejection in a lot that we can accept. If we don't accept single defect in a lot, is the AQL still applicable? I know that normally we won't set AQL to "0" because the sample size is too huge cause the inspection cannot be performed economically and practically, please share your knowledge.


The defect rate does not equal the sample size. If we can assume (safe assumption) that we can tolerate a 0.5% defect rate on average (this is called the AQL), we can develop a sampling plan. Depending upon the maximum percent defective (LTPD) that is acceptable, the sample size is calculated. The following are some examples for c=0

N AQL LTPD
10 0.51% 20.5%
20 0.25% 10.9%
30 0.17% 7.4%
50 0.10% 4.5%
100 0.05% 2.3%

The sample size is independent of the lot size (in general). You can use a standard like Z1.4 that gives you sample sizes that are lot size dependent.
 

Statistical Steven

Statistician
Staff member
Super Moderator
#5
Pangchiapoon -

Acceptable Quality Limit (AQL) is the maximum percent nonconforming that, for purposes of sampling inspection, can be considered satisfactory as a process average. AQL is not lot or batch specific but rather a process average.

Every sampling plan has an AQL. It is impossible to have an AQL of 0.

In statistical terms, the AQL is the percent defective for a given sample size and accept/reject number gives you a probability of acceptance equal to 95%.

Using the sample size, you can also calculate the percent defective that has a 10% probability of acceptance. This is the lot tolerance percent defective (LTPD).

Is that clearer?
 
P

pangchiaboon

#6
Pangchiapoon -

Acceptable Quality Limit (AQL) is the maximum percent nonconforming that, for purposes of sampling inspection, can be considered satisfactory as a process average. AQL is not lot or batch specific but rather a process average.

Every sampling plan has an AQL. It is impossible to have an AQL of 0.

In statistical terms, the AQL is the percent defective for a given sample size and accept/reject number gives you a probability of acceptance equal to 95%.

Using the sample size, you can also calculate the percent defective that has a 10% probability of acceptance. This is the lot tolerance percent defective (LTPD).

Is that clearer?
Thank you Sir.
 
Thread starter Similar threads Forum Replies Date
P Definition Dock-to-Stock Process based on Historical Quality Performance of a Part and Supplier Definitions, Acronyms, Abbreviations and Interpretations Listed Alphabetically 8
M Informational US FDA – Historical Information about Device Emergency Use Authorizations Medical Device and FDA Regulations and Standards News 0
R Need Help on Analysis: How to know potential causality model from historical data Problem Solving, Root Cause Fault and Failure Analysis 5
B Historical Standard Deviation or Select Parts for MSA Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
Ronen E Online resource for USA cities historical weather observations Coffee Break and Water Cooler Discussions 2
K Historical Data to establish Control Limits Manufacturing and Related Processes 4
G Ongoing Cpk and Global/Historical Data Statistical Analysis Tools, Techniques and SPC 7
Marc Atlas of the Historical Geography of the United States Coffee Break and Water Cooler Discussions 2
D Reliability of an Attribute Inspection based on Historical Data Capability, Accuracy and Stability - Processes, Machines, etc. 8
E Predicting Machined Components Capability based on Historical Data of Similar Parts Capability, Accuracy and Stability - Processes, Machines, etc. 1
M Manufacturer's handling of Distributor's Historical Complaint Records US Food and Drug Administration (FDA) 5
B "Historical std deviation" (Standard Deviation) in Gage R&R study Using Minitab Software 6
T Historical Backround of LPA (Layered Process Audit) Process Audits and Layered Process Audits 7
T Six Sigma Register - Historical Body of Evidence Six Sigma 2
B PFMEA Focus - Use of historical data to generate an Occurrence Number FMEA and Control Plans 4
F Documenting Historical Preventive Action at a Daycare Preventive Action and Continuous Improvement 20
M Coming up with a predictor equation for a test using historical data Quality Tools, Improvement and Analysis 1
Hershal DVD and historical movies..... Coffee Break and Water Cooler Discussions 5
S Why historical data & benchmark data is important in APQP planning phase APQP and PPAP 6
G Question on Historical Background on ISO 10012 General Measurement Device and Calibration Topics 2
C SPC: How to set up control limits on SPC chart based on historical data? Statistical Analysis Tools, Techniques and SPC 7
Marc Google Earth in 4D with Historical 'Overlay' Maps After Work and Weekend Discussion Topics 0
T Root Cause Analysis emergency - Historical problems return to plague us! Problem Solving, Root Cause Fault and Failure Analysis 6
M Determination of Product Shelf Life - New Product - No historical data Misc. Quality Assurance and Business Systems Related Topics 10
J Finished Goods Testing - Using historical test data to prove finished goods OK Inspection, Prints (Drawings), Testing, Sampling and Related Topics 5
D How do you structure your QA (QMS) Manual? A Historical Discussion Quality Management System (QMS) Manuals 51
A Information about PPM since giving up AQL - Historical description of requirements AQL - Acceptable Quality Level 12
N Guidance - Cost of Good Quality Misc. Quality Assurance and Business Systems Related Topics 2
D Quality plan for moving locations ISO 13485:2016 - Medical Device Quality Management Systems 3
was named killer Job Opportunity-Quality Engineer-Tampa Florida Job Openings, Consulting and Employment Opportunities 0
Marc Job Opportunity – Quality Assurance Specialist - Suspense 7 April 2021 Job Openings, Consulting and Employment Opportunities 0
C Requirement to link Quality Manual to ISO 9001 clause numbers? ISO 13485:2016 - Medical Device Quality Management Systems 13
normhowe "The Problem with Quality Management: Process orientation, controllability and zero-defect processes as modern myths" Book, Video, Blog and Web Site Reviews and Recommendations 2
J Quality Objective for QMS prior to Certification AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 8
J Quality Objectives and resources ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 11
M PSA Suppliers - CSR matrix and need the quality manual of PSA APQP and PPAP 2
A Quality Control Datasheets Inspection, Prints (Drawings), Testing, Sampling and Related Topics 0
T Quality auditor legal right to see Board meeting minutes ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 12
D Any recommendations on where to find experienced quality hires in Chicago area? Job Openings, Consulting and Employment Opportunities 3
C Budgetary cost to obtain ASME NQA-1 Quality Program Certification Quality Management System (QMS) Manuals 0
O Soft and technical skills for a VP of Quality Career and Occupation Discussions 1
P MedWatch Report from a hospital due to a quality issue US Food and Drug Administration (FDA) 5
Sidney Vianna Release of ISO 10013:2021, Quality management systems – Guidance for documented information Other ISO and International Standards and European Regulations 0
M Unique Quality Management System for 2 sites ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 5
G Application to Chartered Quality Institute - MCQI ASQ, ANAB, UKAS, IAF, IRCA, Exemplar Global and Related Organizations 2
K Quality Agreement MDR - The manufacturer is outside of EU Other Medical Device Related Standards 4
Sidney Vianna Informational APQP4Wind - Advanced Product Quality Planning for the Wind Power Supply Chain APQP and PPAP 3
T 21 CFR 820.20 - Quality Planning Requirements? 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 2
S Supplier protocol for the Quality Supplier Quality Assurance and other Supplier Issues 6
bruceian Software Quality Metrics Software Quality Assurance 11

Similar threads

Top Bottom