Design for Assembly in DFMEA - Failure Effect of Sub-System(s)

[Cephissus]

Hello all,

Another recurring problem that I am facing

For System A to work, it needs Sub-Systems A.1 and A.2 to be assembled together.

In general, a failure of Sub-System A.1 or Sub-System A.2 will lead to a failure of System A.

What happens when a Design failure of Sub-System A.1 will mean that it cannot be assembled to Sub-System A.2 ?? As such System A can't even exist and therefore the discussion about its functions and failures become irrelevant?

How would you describe the Failure Effect of Sub-System A.1 can't be assembled? And how would you rate it in terms of Severity?

 

[Miner]

From your description, you would focus on the severity of the impact to your assembly process. Not knowing your process, we cannot state a specific value. In some processes this severity would be minimal, while in others, it could be pretty severe, resulting in downtime due to the unavailability of parts. I highly doubt that it would be a 9 or 10 as that would mean a process safety issue.

 

[Cephissus]

Ok cool, thanks for the reply. Would you then use the PFMEA Severity tables? To be used inside the DFMEA?

 

[Miner]

If the PFMEA tables take the impact on your process into account, then yes. The usual tables are often very generic and externally focused.

 

[GRP]

From your description, you would focus on the severity of the impact to your assembly process. Not knowing your process, we cannot state a specific value. In some processes this severity would be minimal, while in others, it could be pretty severe, resulting in downtime due to the unavailability of parts. I highly doubt that it would be a 9 or 10 as that would mean a process safety issue.

If you are in the automotive industry, I would add to this the following passage from SAE J1739 related to DFMEA:

The effects of the failure mode should be considered against the next level up assembly, the final product, and the end customer when known. End customer effects should state what the user might notice or experience. They should clearly state if the effect of a failure mode could impact safety or non-compliance to regulations, when applicable. The intent is to forecast the failure effects consistent with the team's level of knowledge.

This means that in the effects column there should coexist effects at multiple levels. It is deemed a good practice to rank the severity of the failure mode as per the highest of the multiple effects associated to it. Also, for a proper severity ranking there should be a collaboration with the customer if possible.

What I find tricky with FMEAs sometimes is how failure modes, effects and causes can mix up depending on how you approach the issue. If you would like to share more of your case, in vague and general terms, we could try to develop it further.

 

[GRP]

Hello all,

Another recurring problem that I am facing

For System A to work, it needs Sub-Systems A.1 and A.2 to be assembled together.

In general, a failure of Sub-System A.1 or Sub-System A.2 will lead to a failure of System A.

What happens when a Design failure of Sub-System A.1 will mean that it cannot be assembled to Sub-System A.2 ?? As such System A can't even exist and therefore the discussion about its functions and failures become irrelevant?

How would you describe the Failure Effect of Sub-System A.1 can't be assembled? And how would you rate it in terms of Severity?

As you say, if the design failure makes it impossible to constitute the product, then you have no function. I would address this as assembly not possible / no function, commonly an 8 in ranking tables, unless safety or govt regulations are involved.