Risk Assessment Precedence - FMEA > Risk Matrix (Modified PHA) > Ishikawa?

v9991

Trusted Information Resource
at a design/process development phase,
of the two tools, PHA(risk matrix) & Ishikawa, which one comes first?
I believe that, FMEA would be used finally to nail-down various modes-identified risks/failures.

(I guess, that answer could be, FMEA alone is adequate, but to enable a large number of teams to independently adopt risk assessment process, what do you suggest)
 

Marcelo

Inactive Registered Visitor
Usually it depends on the stage of the design phase. No, FMEA does not work at all for everything as people usually thinks.

FMEA was created to be used when you have a defined product with defined components, because if focuses on component failure. A good time to use it is after an initial prototype is created. It?s base is cause-effect.

PHA is a technique that, as the name implies, can be used in initial stages.

Ishikawa is a technique where you begin with effect and you find causes, so it can also be used in the initial stages when you know problems but not exactly have a detailed design.

You may want to take a look at IEC 31010 - Risk management ? Risk assessment techniques, which has good explanations and ways to use several techniques.
 

v9991

Trusted Information Resource
We are suggesting similar fashion.
Ishikawa ==> risk matrix (modified PHA) ==> FMEA --> development experimentation/trials ==> updated PHA & FMEA ==> Control strategy


we have been driving to link the experimentation trials with the 'FMEA-current controls and action plans'.

but there is a opinion among the team, that design FMEA would not be relevant (or non-value adding) because it is early stages and much of the expertize/knowledge may-not-be-specific to the given product.
+ PHA could be adequate to support QRA requirement of agencies and also reflect the RA of the product/process!

1) how do we counter that argument!
2) can you also please confirm that ishikawa leads to risk matrix (/ PHA)! (my opinion is that if PHA leads to ishikawa representation, it would be only a retrospective/representation of the risk assessment RA, )
 

Marcelo

Inactive Registered Visitor
We are suggesting similar fashion.
Ishikawa ==> risk matrix (modified PHA) ==> FMEA --> development experimentation/trials ==> updated PHA & FMEA ==> Control strategy


we have been driving to link the experimentation trials with the 'FMEA-current controls and action plans'.

but there is a opinion among the team, that design FMEA would not be relevant (or non-value adding) because it is early stages and much of the expertize/knowledge may-not-be-specific to the given product.
+ PHA could be adequate to support QRA requirement of agencies and also reflect the RA of the product/process!

1) how do we counter that argument!
2) can you also please confirm that ishikawa leads to risk matrix (/ PHA)! (my opinion is that if PHA leads to ishikawa representation, it would be only a retrospective/representation of the risk assessment RA, )

I really am puzzled at your comments. For example, you second question - ishikawa leads to risk matrix (/ PHA). Both Ishikawa and PHA are techniques, they lad to nothing in principle.

I think, to try to help you, it would be better to understand, what are you exactly trying to do?
 

v9991

Trusted Information Resource
I really am puzzled at your comments. For example, you second question - ishikawa leads to risk matrix (/ PHA). Both Ishikawa and PHA are techniques, they lad to nothing in principle.

I think, to try to help you, it would be better to understand, what are you exactly trying to do?

Sorry, for not being specific/elaborating my query. (when i said ' Ishikawa leading to PHA', I meant that usage/utilization of the tool during the development process)

I meant to convey that, while doing a Quality Risk Assessment, first we would take each of the 'key-feature or requirements' and do a brain storming session to complete the ishikawa diagram.

Then summarize them, the 'positive or contributing' & ' negative or constraining' factors through a risk matrix (modified PHA).

at the end of development, this risk matrix & FMEA will be used to reflect the overall risk (closed vs residual).
 

Marcelo

Inactive Registered Visitor
By risk matrix, do you mean something like the attached example? This is usually called a risk matrix.

I still think you are confusing the risk assessment (or is it risk management that you want to perform?) process with tools used to gather information throughout the process. Please take a look at the other attachment, it details a generic risk management process. Risk assessment is a part of the process. Tools should be used during the process (from what I understand your question may be related on how to use the tools along the process?). A risk matrix is usually part of the risk evaluation, because it?s a decision too.
 

Attachments

  • risk matrix.jpg
    risk matrix.jpg
    74.3 KB · Views: 1,575
  • Quality risk management.png
    Quality risk management.png
    21 KB · Views: 842

v9991

Trusted Information Resource
Tools should be used during the process (from what I understand your question may be related on how to use the tools along the process?). A risk matrix is usually part of the risk evaluation, because it?s a decision too.

yes, you are right, I meant the usage of tools during risk management process. we are trying to outline the tools to be employed during the product development process.


By risk matrix, do you mean something like the attached example? This is usually called a risk matrix.

by risk matrix i meant a an matrix assessing the "process parameters vs quality attributes". and this summary would come-up from brain storming session. (instead of directly jumping directly into this assessment, we are trying to recommend to complete a Ishikawa on each of quality attribute / requirement) (Again, sorry for the mix up of terminologies, )
 

Marcelo

Inactive Registered Visitor
Ah ok, you mean something like the attached then.

by risk matrix i meant a an matrix assessing the "process parameters vs quality attributes". and this summary would come-up from brain storming session. (instead of directly jumping directly into this assessment, we are trying to recommend to complete a Ishikawa on each of quality attribute / requirement) (Again, sorry for the mix up of terminologies, )

Normally, you use brainstorming or Ishikawa to identify the parameters or variables. Then you use a tool such as FMEA (to tell you the truth in this case you should use FMECA due to criticality) to rank the risks. You can?t use Ishikawa for this because it?s not a tool to rank the risk.

You would then have what you call the risk matrix. You may take a loo at this article for some explanations and examples: http://scholarsresearchlibrary.com/DPL-vol2-iss2/DPL-2010-2-2-370-392.pdf
 

Attachments

  • Screen Shot 2014-09-17 at 22.59.22.png
    Screen Shot 2014-09-17 at 22.59.22.png
    74.6 KB · Views: 468

v9991

Trusted Information Resource
Do you suggest that modified PHA & FMEA to be used in the process; like i said in my earlier response,
* PHA would summarize and represent the overall view of "PP vs QAs",
* FMEA would drive the action plans and residual risk!! and especially link with the control strategies


Thank you very much for patience & confirmation (and your patience )
 

Marcelo

Inactive Registered Visitor
Another question, by PHA you mean Process Hazard Analysis or Preliminary Hazard Analysis? I think you mean process (however, the acronym is mot used for preliminary :))

Preliminary Hazard Analysis is a tool, such as FMEA and FTA.

Process Hazard Analysis is more a "type" of risk management process (originally from the chemical industry), but if differs from quality risk management as defined in the ICH documents. For example, Process Hazard Analysis expects the use of the same tools such as Ishikawa, FMEA, FTA, etc.


* PHA would summarize and represent the overall view of "PP vs QAs",
* FMEA would drive the action plans and residual risk!! and especially link with the control strategies

Again I would suggest that you defined more clearly what you want in each step. For example:

Step 1 - Determine Critical Quality Attributes - you can use Brainstorming, QFD, etc, any tool that enables you to determine the attributes of the product
Step 2 - Identify process parameters - you could use Ishikawa or any other tool to identify those
Step 3 - Link process parameters with CQA and risk rank then - you can use FMECA or any tool that enables assessment and ranking
Step 4 - Identify control measures - there?s a lot of different tools for that, FMECA being one.
 
Top Bottom