Special Inspection Levels - Batch sizes are large (typically 100k - 150k)

S

shbee

#1
Hi

This is my first post, but I could do with some help from a wider audience than I have available to me in my company please.

First some background.....
We have a new process manufacturing a new product. Batch sizes are large (typically 100k - 150k). One of the acceptance criteria can only be measured when the product is ready for shipment and is a variable.
We are planning to use variable AQL based sampling plan using ISO6002 (which I believe is comparable to MIL-STD-105E and ANSI/ASQ Z1.4). The AQL for the characteristic is 1.0%.
I've put together a sampling plan using general inspection level II as this is a new process, however the sample size is coming out as 150. This doesn't sound a lot on a 150,000 batch size but it will take at least 1 week to test this. We are producing approx 2 batches / week so this isn't really viable in the long term.
One other complication is that the test method is a destructive test and 150 samples has a significant cost.

Now to the questions....
Under what circumstances can the special inspection levels S-1, S-2 etc be used? I've read that they offer a lower sampling size and are especially useful for destructive and expensive tests.
If special levels are used, then surely some of the confidence in the result must be lost, so what are the implications of using a special inspection level?
Also, if I do chose to use a special level, how can I determine which one is the most appropriate?

Many thanks in advance.
 
Elsmar Forum Sponsor
P

Pudge 72

#2
Re: Special Inspection Levels

Is this a single component with a specific dimension, or Final measurment of an Assembly / Assembled Product?
 
S

shbee

#3
This is a single measurement of a finished product which will be shipped directly to the consumer. The product is a type of fabric coated with a gel. The measurement is not a dimension, but a measure of how 'tacky' or 'sticky' the gel on the product is.
Thanks
 
P

Pudge 72

#4
I guess my next question would be - have you performed Attribute Gage R&R whereas you have defined the criteria of "Sticky" or "Not Sticky" with your Inspection department or Operators?
After that, I would review what your failure rate was during development / engineering in combination with the results of your R&R?
In other words, do you have a known Defect Rate / Benchmark for which to formulate the sample size on, or is what you are basing your AQL Level on a best guess as to what the failure rate could be? Is there justification for an assigned level - or no?
I am trying to understand if we can tie the engineering work that has been done to what you are trying to accomplish in production, or if more leg work needs to be done. Unfortunately - everyone would like to get rid of desructive testing, but if it is not possible (even if it is expensive), the best support that I have found is the benchmark data to associate the frequency and sampling size with when assigning the AQL.
 
S

shbee

#5
Thanks for your responses Pudge 72.

The measurement we are taking is not an attribute, but a variable force measurement. The test method is a good repeatable method, which we have data to support.
Our AQL level is based primarily on the classification of the impact of the defect on our customer - at this stage in our development, the AQL is pretty much fixed and I'd struggle justify changing it.
In any case as I understand ISO 6002, the sample size is not dependant on the AQL, but the batch size & inspection level. (It's the acceptability constant that is dependant on the AQL).
My batch size is fixed, hence I'm looking to reduce the inspection level if possible.
 
P

Pudge 72

#6
I will bow out here as I am completely unfamiliar with the 6002 Standard and its requirements.

My last gasp would be to try and justify reducing sampling based on known defect levels, or in this case even better - associated capability levels from the inspection of the feature & tie that to Sigma Levels and PPM as your justification. Developing a rationale based on known production quantities in accordance with statistically proven capability levels should be a sound plan. In my opinion, a sampling plan based on a confidence interval driven from your data would be the best route to go.
But, I am not a statistician either :D.

Very intersting topic - sorry I could not be of more help.
 
S

shbee

#7
Thanks very much for your help Pudge 72. :thanx:

Is there anyone else who may be able to help me regarding when and which Special Inspection level I can use during AQL sampling?
 
T

Tavoludo

#8
Hi Shbee.
Do you already fix your problem? The reason why some inspection need to be done on finished products is because we do not have controlled processes. So base on that, insted of answer your question I recomend you to work with some statistical procees control on that process you have, specifically you can use a Pre-control charts and split your production in small lots (Coluld be hourly) and use you statistical results to take the desition of release or not that specific batch of product.
 

Ronen E

Problem Solver
Staff member
Moderator
#9
Hi

This is my first post, but I could do with some help from a wider audience than I have available to me in my company please.

First some background.....
We have a new process manufacturing a new product. Batch sizes are large (typically 100k - 150k). One of the acceptance criteria can only be measured when the product is ready for shipment and is a variable.
We are planning to use variable AQL based sampling plan using ISO6002 (which I believe is comparable to MIL-STD-105E and ANSI/ASQ Z1.4). The AQL for the characteristic is 1.0%.
I've put together a sampling plan using general inspection level II as this is a new process, however the sample size is coming out as 150. This doesn't sound a lot on a 150,000 batch size but it will take at least 1 week to test this. We are producing approx 2 batches / week so this isn't really viable in the long term.
One other complication is that the test method is a destructive test and 150 samples has a significant cost.

Now to the questions....
Under what circumstances can the special inspection levels S-1, S-2 etc be used? I've read that they offer a lower sampling size and are especially useful for destructive and expensive tests.
If special levels are used, then surely some of the confidence in the result must be lost, so what are the implications of using a special inspection level?
Also, if I do chose to use a special level, how can I determine which one is the most appropriate?

Many thanks in advance.
Hi and welcome to the cove :bigwave:

1. In my opinion, if you have variable test data it's a shame to go by an attribute sampling plan altogether. Variable data will give much more insight than attribute data on the same sample size. A sample size of 30 variable data points would allow in many cases a good distribution fit and hence a sound confidence interval could be established. If you are concerned with destructive testing costs consider applying the Bootstrap technique to improve your confidence with smaller sample sizes.

2. Re. your other question on reduced confidence with smaller sample sizes - most attribute sampling plan standards (e.g. AS 1199) provide graphic and tabulated information to work out the confidence level associated with most specific plans described in the standard. Look at the back pages...:lol:

Cheers,
Ronen.
 

Ronen E

Problem Solver
Staff member
Moderator
#10
Hi Shbee.
Do you already fix your problem? The reason why some inspection need to be done on finished products is because we do not have controlled processes. So base on that, insted of answer your question I recomend you to work with some statistical procees control on that process you have, specifically you can use a Pre-control charts and split your production in small lots (Coluld be hourly) and use you statistical results to take the desition of release or not that specific batch of product.
I'd second that, but please note that the critical parameter can only be measured once the product is complete, hence the relationship between the control process parameters and the final critical parameter must first be established, to allow the method you suggested to be effective and reliable.

Further, I would suggest validating the process, so that once the critical process parameters are kept within limits, an acceptable outcome is guaranteed (or at least assured to a known confidence level). This way FPT can be eliminated altogether or at least reduced to a low monitoring level (so action can be triggered once alert limits are breached).

Cheers,
Ronen.
 
Thread starter Similar threads Forum Replies Date
M Basis for the Derivation of Inspection Levels (I, II, III & Special) AQL - Acceptable Quality Level 8
D What Inspection Level for Small Lot Size and Special Levels (S1-S4)? Inspection, Prints (Drawings), Testing, Sampling and Related Topics 15
Y What are different Special Inspection Level 1-4 and General spesification 1-3 ? AQL - Acceptable Quality Level 0
S How to incorporate a customers special inspection frequency in our control plan FMEA and Control Plans 1
G Special Inspection vs. General Inspection AQL - Acceptable Quality Level 7
T Special Inspection Level 3 (S-3) per ANSI Z 1.4 & its relationship with ISO 2859-1 AQL - Acceptable Quality Level 1
Techilady Ford Special Characteristics Communication and Agreement Form (SCCAF) provides production requirements? FMEA and Control Plans 4
R Critical Characteristics & Special Characteristics FMEA and Control Plans 2
P Letter to file or special 510k or traditional 510k 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 2
S OEM full range calibrations vs Validation special test points Medical Device and FDA Regulations and Standards News 0
G Special 510K possible? Only the packaging has changed Other US Medical Device Regulations 0
S Special 510(k) for orthopedic implant 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 4
N Riveting - special process Manufacturing and Related Processes 11
H AS9100D clause 8.5.1 f) & 8.5.1 g) - Special Processes AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 2
C AIAG-VDA PFMEA - 1st special Characteristics? [5step vs. 6step] FMEA and Control Plans 3
J 1.11 Preliminary Identification of Special Product and Process Characteristics APQP and PPAP 4
earl62 What is the best way to control special characteristics in Control plan? Is it Mandatory to have SPC for IATF 16949? IATF 16949 - Automotive Quality Systems Standard 12
L Special Processes for SERVICES AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 10
K PSA Special characteristics Customer and Company Specific Requirements 0
Watchcat Advantages to pursuing a Special 510(k) instead of a Traditional 510(k) Other US Medical Device Regulations 5
qualprod Managing a "special" customer into the QMS? ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 10
S Special 510(K) - Labeling change from Rx to OTC Other US Medical Device Regulations 11
M Special Controls and Class II Medical Devices Medical Device and FDA Regulations and Standards News 18
A ISO 10993-7:2008/ Amd.1:2019 - Classification of special populations Other Medical Device Related Standards 3
H SPC for special characteristics - CSR - Control of SC and CC Statistical Analysis Tools, Techniques and SPC 1
L Special Processes - Technical reason concerning torque - Anaerobic compounds Manufacturing and Related Processes 10
M Informational US FDA Final Guidance – Special Premarket Notification [510(k)] Pathway Medical Device and FDA Regulations and Standards News 0
F Replacement Bone Screws - Traditional 510(k) or Special 510(k)? Other US Medical Device Regulations 7
F Assignment of Special Characteristics to government regulations IATF 16949 - Automotive Quality Systems Standard 5
qualprod ISO 9001:2015 8.5.1 f - Scope with customer special process requirements ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 11
J ISO 13485 - 7.5.2 cleanliness , 6.4.2 validation of special processes, and 6.4.2 contamination ability to be exempt? ISO 13485:2016 - Medical Device Quality Management Systems 5
L Injection Moulding Process (Special Process) Validation Food Safety - ISO 22000, HACCP (21 CFR 120) 2
S The key differences in 510k types - Traditional vs. Abbreviated vs. Special US Food and Drug Administration (FDA) 0
D I'm new in IMDS special using the IMDS application Manufacturing and Related Processes 8
S VW Special status condition IATF 16949 - Automotive Quality Systems Standard 3
K Letter to file vs Special 510 for additional accessory in orthopaedic kit Other US Medical Device Regulations 1
P Special 510(k) - Modified Medical Device Description Other US Medical Device Regulations 8
P V&V for Special 510k 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 2
F CQI-23 Special Process Molding - Controls IATF 16949 - Automotive Quality Systems Standard 7
B AS9100 - Are special process suppliers required to be NADCAP accredited? AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 15
A Non-destructive testing methods are special processes? Inspection, Prints (Drawings), Testing, Sampling and Related Topics 18
P Sanctioned Interpretations - IATF 16949 Cl. 8.3.3.3 Special characteristics IATF 16949 - Automotive Quality Systems Standard 3
E IATF 16949 Cl. 8.2.3.1.2 - Customer-designated special characteristics IATF 16949 - Automotive Quality Systems Standard 1
T Customer special status and initial IATF certification IATF 16949 - Automotive Quality Systems Standard 12
J NADCAP - Supplier Special Processes Certifications Manufacturing and Related Processes 1
N Special Vs Non-Special Processes Misc. Quality Assurance and Business Systems Related Topics 15
D Customer-Designated Special Characteristics vs. Special Characteristics IATF 16949 - Automotive Quality Systems Standard 2
S IATF 16949 Clause 8.3.3.3 - Documentation of all Special Characteristics in Drawings IATF 16949 - Automotive Quality Systems Standard 2
A Special Procedures for Non Aerospace Customers AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 3
M AS9110C Clause 8.5.1.2 Validation and Control of Special Processes for Repair Station AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 5

Similar threads

Top Bottom