Probability of Occurrence Ranking for Low Volume Manufacturing vs High Volume

K

kennethshaw

#1
Hi,

I'm a newbie to the site. I'm currently working on post market surveillance ranking scales for a global company that manufactures any where from low to very high volume products. I was wondering if anyone could tell me how to scale the probability of defect occurring based on the quantity of production. I've heard that it is acceptable to scale the ranking based on quantity. If so, is there a justification or best practice on how this is done?

Thanks.

Ken
 
Elsmar Forum Sponsor

Chennaiite

Never-say-die
Trusted Information Resource
#4
Hi,

I'm a newbie to the site. I'm currently working on post market surveillance ranking scales for a global company that manufactures any where from low to very high volume products. I was wondering if anyone could tell me how to scale the probability of defect occurring based on the quantity of production. I've heard that it is acceptable to scale the ranking based on quantity. If so, is there a justification or best practice on how this is done?

Thanks.

Ken
AIAG refers category of occurrence rating in subjective terms as Very High, High, Moderate, Low and Very Low. To be honest, more often than not, while doing an FMEA, this is the one I effectively use. Important thing is to ensure that the arrived rating is consistent with the actual rejection figures and also the robustness of the preventive controls in place. If you still want to pursue a rating table that relates to only 'numbers', one option I would suggest is to customize based on rejection targets.
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#5
there may be some confusion:

The typical use of 'occurence' is the defect rate (in either % or ppm) which takes into account the volume.

However severity of effect can be affected by the actual number of occurences.

I will add that strictly speaking the phrase "probability of occurence" dosn't relate to defect rate, it refers more to the liklihood that a defect or failure will occur. This phrase is more suited to DESIGN FMEAs and other risk assessments...
 
K

kennethshaw

#6
Thanks for your comments all. Maybe it might help to give a little more details.

1. I work for a medical device company that makes anywhere from 100 devices to 100,000 devices per year.
2. Our post market risk analysis ranks the following:
2a. Probability of occurrence (this is based off of how much we produce over certain date ranges and how many failures have been reported. This is a numerical scale that is based on frequency ranges.
2b. Probability of injury occurring if defect is not detected.
2c. Severity of injury.

My gut is saying that for score 2a (probability of occurrence), it shouldn't make a difference what the production quantities are when I'm focusing on the affect on the patient. In other words, 5 defects out of 50 devices produced represents a higher risk to the patient than 5 defects out of 100,000 devices produced.

Hope this helps generate more discussion. I'm at a loss since literature searches are coming up dry. I'm brainstorming what factors play a part in developing a custom scale for the range of products and I would think the following would apply: Criticality of the product (band-aid vs artificial heart), AQL level. Anything else I can consider? What I want to do is keep business risk out of it.

Thanks in advance for your help.

Ken
 
K

kennethshaw

#7
there may be some confusion:


I will add that strictly speaking the phrase "probability of occurence" dosn't relate to defect rate, it refers more to the liklihood that a defect or failure will occur. This phrase is more suited to DESIGN FMEAs and other risk assessments...

Unfortunately, I'm more confused now after reading your statement. Isn't defect rate a measure or a way to predict how likely a defect or failure will occur?
 

Chennaiite

Never-say-die
Trusted Information Resource
#8
My gut is saying that for score 2a (probability of occurrence), it shouldn't make a difference what the production quantities are when I'm focusing on the affect on the patient. In other words, 5 defects out of 50 devices produced represents a higher risk to the patient than 5 defects out of 100,000 devices produced.

Ken
I have two school of thoughts here. Sometimes(going by layman's statistics), defects from low volume do not really indicate the likelihood of occurrence, just like a two-match series does not 'confidently' indicate the likelihood that a team will win or lose in the subsequent encounters - perhaps a ten-match series does. I would also agree that taking low volume lightly will deflate the risk projection at times.

In the end, the best possible option is perhaps to stick to the judgement based on subjective terms. Of course, a definite customized rating table will help in consistent judgement across your organization. This can be arrived based on studying how the inarguably best FMEA team of your organization make judgements in real scenario.
 
Last edited:

Bev D

Heretical Statistician
Staff member
Super Moderator
#9
Ken I think you are over thinking this. Can you describe in more detail what you are trying to do and what actions your analysis might generate?
 

Chennaiite

Never-say-die
Trusted Information Resource
#10
..I will add that strictly speaking the phrase "probability of occurence" dosn't relate to defect rate, it refers more to the liklihood that a defect or failure will occur. This phrase is more suited to DESIGN FMEAs and other risk assessments...
(Similar product)Defect rate along with Preventive controls determine liklihood, don't they.

..Can you describe in more detail what you are trying to do and what actions your analysis might generate?
I share this query too, especially since it is stated as 'Post Market Risk analysis'. Risk analysis as far as I understand helps in making decision. I am not from Medical field, but what decision is made by analyzing risk post-market, unless otherwise it is to do with development device.
 
Last edited:
Thread starter Similar threads Forum Replies Date
H Risk Analysis and Probability of Occurrence ISO 14971 - Medical Device Risk Management 3
Sravan Manchikanti Software Risk Management & probability of occurrence as per IEC 62304 IEC 62304 - Medical Device Software Life Cycle Processes 8
N Hazard Probability of Occurrence Definition ISO 14971 - Medical Device Risk Management 11
E Risk Management selection Probability of Occurrence and Severity ISO 14971 - Medical Device Risk Management 24
C Looking for a table that quantifies Likelihood/Probability of occurrence of a hazard ISO 13485:2016 - Medical Device Quality Management Systems 3
M Should volume of sales be factored into risk probability assessments? ISO 14971 - Medical Device Risk Management 33
MrTetris How to qualitatively assess initial probability ISO 14971 - Medical Device Risk Management 4
R Probability - Need a help to solve the below question Statistical Analysis Tools, Techniques and SPC 5
Ronen E Tolerance intervals (?) question - Flow Rate Probability Range Statistical Analysis Tools, Techniques and SPC 6
K Small Sample - Lower Limit Probability Statistical Analysis Tools, Techniques and SPC 0
Y IEC 62304 Section 4.3(a) - 100% probability of failure IEC 62304 - Medical Device Software Life Cycle Processes 3
L Balls drawn without replacement probability question Brain Teasers and Puzzles 8
L How to calculate the Probability of Deficiency of the AQL of an Inspection AQL - Acceptable Quality Level 16
Marc Definition PFH - Probability of dangerous failure per hour Definitions, Acronyms, Abbreviations and Interpretations Listed Alphabetically 0
Marc Definition PFD - Probability of failure on demand Definitions, Acronyms, Abbreviations and Interpretations Listed Alphabetically 0
A Risk Acceptability Criteria - Probability and Acceptability Level ISO 14971 - Medical Device Risk Management 1
M Risk Probability - Is there a Better System? ISO 14971 - Medical Device Risk Management 18
J Can Normal Probability Paper and GPC be created with Excel Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
Z What is the Probability of Picking Up "Good" alphabets? Six Sigma 6
C How to calculate P value on Probability Plot Statistical Analysis Tools, Techniques and SPC 1
Bev D Probability question - Components of Variation study on a single use diagnostic test Statistical Analysis Tools, Techniques and SPC 1
P One sided confidence interval for probability plot in Minitab Using Minitab Software 1
T Repeated Inspection - Probability of Visual Inspection catching Defects Inspection, Prints (Drawings), Testing, Sampling and Related Topics 2
L "Half Normal Plot vs Normal Probability Plot" of Effects Using Minitab Software 1
U PROBLEM: Probability of Resource Access Collision Statistical Analysis Tools, Techniques and SPC 5
S Software Risk Estimation: Probability of Medical Device Software Anomaly Occuring ISO 14971 - Medical Device Risk Management 9
N Bayes Law of Conditional Probability ("The Drunkard's Walk" by Mlodinow) Statistical Analysis Tools, Techniques and SPC 1
S Probability Density Function of a Function of a Random Variable Statistical Analysis Tools, Techniques and SPC 13
V Preparing for SSGB - Six Sigma Probability Question Six Sigma 13
R Risk Management - Probability vs. Frequency IEC 60601 - Medical Electrical Equipment Safety Standards Series 3
V ASQ CSSGB - More In Depth Probability and Sampling Methods prep materials Book, Video, Blog and Web Site Reviews and Recommendations 4
A Probability of a good lot given that the sample was good Inspection, Prints (Drawings), Testing, Sampling and Related Topics 8
Rameshwar25 How Probability of Acceptance of a Part is Calculated Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
B Plotting Probability Exceedance using the Probability Plot Graph in Minitab Using Minitab Software 4
I Mathematical Probability Puzzle Brain Teasers and Puzzles 25
1 What are TOP, DPU, DPO, DPMO? Mult-CTQ (discrete) ZST - Is this a probability? Six Sigma 1
B Probability of finding a Defective Part - Probability question Statistical Analysis Tools, Techniques and SPC 8
D ISO 2859 - Probability that sample size is sufficient Inspection, Prints (Drawings), Testing, Sampling and Related Topics 2
C Multiple Censored Tests to find out the Survival Probability Reliability Analysis - Predictions, Testing and Standards 3
M Probability Distribution Problem Inspection, Prints (Drawings), Testing, Sampling and Related Topics 5
M Confidence Interval on Normal Probability Plot Using Minitab Software 2
bobdoering How's this for a probability function? Player Hits Own Mom With Foul Ball Funny Stuff - Jokes and Humour 7
D Probability of Failure - Seven expansion joints Reliability Analysis - Predictions, Testing and Standards 5
bobdoering Goodness of Fit: Is it normal? Probably...or probably not - it's all probability Imported Legacy Blogs 0
J Xbar - R Control Chart taking into account the Probability Limit Using Minitab Software 8
0 Calculating axis values of normal probability plot Statistical Analysis Tools, Techniques and SPC 4
R Sampling - Total probability of accepting bad lot - Producer's risk Inspection, Prints (Drawings), Testing, Sampling and Related Topics 2
W Interpreting a Normal Probability Plot Statistical Analysis Tools, Techniques and SPC 4
A Actual probability of TYPE I error for 3 sigma P-chart Statistical Analysis Tools, Techniques and SPC 10
G Running simulations using Bernoulli Distributions with a certain probability. Using Minitab Software 5

Similar threads

Top Bottom