How to Reduce Sampling and get the same MTTF (Mean Time to Faliure)

D

DomDom

#1
Good day all,

I am working on a subject in regards to MTTF. So far, our standard work asks us to provide 36 units to put them in life test. this test is to determine the MTTF of the lot. This is for the new product development.

36 units is a large number, especially when the test is destructive and we are still in development.

I would like to find a way to demonstrate, right or wrong, that we can reduce the number of sample. Ideally it will be better for us to test 10 units. This will allow the development to try more configuration at iso-cost before getting deeper with the best configuration.

So, my question is I hope simple. Is there a way, with minitab or by anyother mean, to show that reducing the samples from 36 to 10 will give a similar result with a similar 95% CI? (The value 10 is arbitrary at this stage)

Thank you for your help on this problem.

Regards,
DomDom
 
Elsmar Forum Sponsor
B

Barbara B

#2
DomDom,

with what method did you get a result of 36 samples? Could you pls give us some more information?

In general: If you test less than the number of pieces required (e.g. with Stat > Reliability/Survival > Test Plans > Demonstrate) under the same conditions which were used to calculate this sample size, you'll get a broader CI because the reliability of information is smaller with 10 parts than with 36.

Regards,

Barbara
 

Miner

Forum Moderator
Leader
Admin
#3
In addition to what Barbara stated, the only way you can maintain the same confidence level while reducing the number of samples would be to extend the demonstration testing time, usually significantly.

Using Stat > Reliability/Survival > Test Plans > Demonstrate, you can enter both 36 and 10 (or 36 30 25 20 15 10) into the sample size field. Leave test time blank. The Session Window will show the test duration next to each sample size while maintaining the 95% confidence level.
 
D

DomDom

#4
Thank you for the comments.

I had a thought this night. In fact, what we want to do is to be bale to choose from 3-4 configuration.
Before we needed 36 (number chosen by our R&D people based on the minimum population to have a correct Cp, Cpk) samples, for each config.
Instead of having this, I will take these 3-4 config and only 10 sample for each config. I will then do comparison between them to understand which one is the best.
After that, I believe we still need to have a large population to have a good estimate of survival rate for the selected config.

As our products do not have any history (tailor made products) we cannot assume any distribution before we do the tests.

So, what do you think about the process stated above?

:thanx: for your feedback,
DomDom
 
B

Barbara B

#5
DomDom,

to compare different configurations you could take an arbitrary number of parts (e.g. 10) for each configuration and analyze the data with a methode called "regression with life data". There you can get informations about the reliability of the different configurations.

You can also take the study design one step further if the configurations can be described by several factors (like process time, material used, etc.) and these factors can be set up deliberately. Then the configurations are made up by a statistical approach and not by the engineers, they provide only the range for the different factors (region of interest). With this setup you can conduct a DoE and analyze the impact of factor settings on the reliability (design: factorial doe, method for analysis: regression with life data).

Either way you can evaluate which distribution is appropriate for the reliability data in the different configurations. For life data a Weibull distribution is a common one, but there are others in the field of reliability/survival which could be used (like Exponential, Lognormal, or others).

Regards,

Barbara
 
Thread starter Similar threads Forum Replies Date
D Project to Reduce Destructive Sampling in Crimping Process Inspection, Prints (Drawings), Testing, Sampling and Related Topics 3
Y Can I reduce the Sample Size? Sampling check of more than 1,000 pieces daily Inspection, Prints (Drawings), Testing, Sampling and Related Topics 25
D When do you reduce SPC sampling? And How? Inspection, Prints (Drawings), Testing, Sampling and Related Topics 7
V Sample Size rationale to reduce size and use special levels EU Medical Device Regulations 6
PQ Systems 5 Ways to Reduce Stress on Your Next Audit Using GAGEpack Software 0
C Can I reduce P1 to zero? ISO 14971 - Medical Device Risk Management 15
L Reduce risks as far as possible - Quartz Crystal ISO 14971 - Medical Device Risk Management 11
S Advice on how to reduce overhead of handling non-conforming material Nonconformance and Corrective Action 7
M Reduce occurrence rating based on the PMS data and customer complaint data ISO 14971 - Medical Device Risk Management 2
C Stress / Challenge Conditions for Design Verification Testing to Reduce Sample Size 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 11
A How to reduce the process SPC monitoring Capability, Accuracy and Stability - Processes, Machines, etc. 3
adir88 Information of safety can reduce risk now? ISO 14971 - Medical Device Risk Management 12
Ashok sunder Is it possible to reduce Risk likelihood and impact Post control Ranking after corrective action taken for risk? FMEA and Control Plans 1
Ashok sunder Is it possible to reduce FMEA Occurrence and Detection Ranking after corrective action taken for customer complaints? FMEA and Control Plans 6
W Duplicated Gauges want to reduce calibration General Measurement Device and Calibration Topics 2
S Methods to reduce relative humidity in Datacenter IEC 27001 - Information Security Management Systems (ISMS) 3
O How do reduce the risk of suppliers having similar problems? Supplier Quality Assurance and other Supplier Issues 4
R Employee diet to reduce absenteeism and increase productivity Hospitals, Clinics & other Health Care Providers 24
B Proactive efforts to reduce risk - PFMEA risk reduction activities IATF 16949 - Automotive Quality Systems Standard 8
N How to reduce RPN of Visual Inspection Process FMEA and Control Plans 6
2 High-tech medical equipment to reduce medication errors Medical Information Technology, Medical Software and Health Informatics 5
M Is That Plating Wave Washer will reduce its stiffness? Supplier Quality Assurance and other Supplier Issues 1
C How to Reduce Inspection Frequency in Metal Stamping Company Quality Manager and Management Related Issues 4
C Design of Experiments to Reduce Variation in a Lathe Operation Quality Assurance and Compliance Software Tools and Solutions 9
D How to reduce the number of Quality Inspectors Quality Manager and Management Related Issues 7
S Using ANSI/ASQ Z1.4 to Reduce Impact of Field Service Campaign AQL - Acceptable Quality Level 17
P How to reduce or shut off Compressed Air for Painting Robots Manufacturing and Related Processes 3
P How to Reduce or Eliminate Inspection on an Assembly Line Quality Tools, Improvement and Analysis 10
Sidney Vianna ISO 14006 - New Standard to Reduce Environmental Impacts of Products and Services Miscellaneous Environmental Standards and EMS Related Discussions 1
J Reduce 50 Work Instructions into 12 Key Processes Process Maps, Process Mapping and Turtle Diagrams 8
A Can we reduce documents checking percentage on supplier certificates? Supplier Quality Assurance and other Supplier Issues 3
T How do I Reduce Frequency of Receiving Inspection Testing for Quality Control Inspection, Prints (Drawings), Testing, Sampling and Related Topics 9
A PFMEA (Process FMEA) Detection Rating - Actions to Reduce RPN FMEA and Control Plans 3
S What to do if no further control possible to reduce the OHS risk? Occupational Health & Safety Management Standards 16
S How to Reduce the Cost of Broken Measurement Equipment Quality Tools, Improvement and Analysis 8
M Plastic Mold Approval - What are the ways to reduce mold approval cycle? Manufacturing and Related Processes 8
A How to Reduce Printed Paper Waste Sustainability, Green Initiatives and Ecology 47
P Does Six Sigma Project have to Reduce Variability? What about a mean shift to target? Six Sigma 6
M Does Disciplinary Action Reduce Operator Errors? Misc. Quality Assurance and Business Systems Related Topics 80
A Control Chart to Track Effectiveness of Actions to Reduce Scrap Rate Six Sigma 7
P Attempting to reduce sample size - The "best" statistical technique to use Statistical Analysis Tools, Techniques and SPC 10
Sidney Vianna ISO Guide 64 will help reduce environmental impacts of products Other ISO and International Standards and European Regulations 1
B On Time Delivery Question - Project to reduce late shipments Lean in Manufacturing and Service Industries 7
T Using SPC Control Charts to Reduce AQL Sample Size AQL - Acceptable Quality Level 16
R CMM Machine upgrade - Ways to reduce CMM inspection cycle time Inspection, Prints (Drawings), Testing, Sampling and Related Topics 5
J Customer wants to Reduce GD&T Tolerances on Sheet Metal Subassemblies Manufacturing and Related Processes 8
Douglas E. Purdy Classification of Suppliers & Control Plans to Reduce Receiving Inspection FMEA and Control Plans 6
R Tool Kits to Reduce Direct Time on Milling Machine Manufacturing and Related Processes 15
J Cost Down! - Reduce Manpower Misc. Quality Assurance and Business Systems Related Topics 7
R 100% Visual Inspection - Would you Remove or Reduce the Visual Inspection? Inspection, Prints (Drawings), Testing, Sampling and Related Topics 9

Similar threads

Top Bottom