What's special about your RCAs (Root Cause Analysis)?

I

ibillwilson

Hi All,

I approach Root Cause Analysis (RCA) in a particular way due to my industry. I'm trying to learn more about how others do RCA. So, I'm curious... what makes the way your industry does RCA different?


  1. What industry or domain is the setting for your RCA efforts?
  2. For what kinds of issues do you perform RCAs?
  3. What makes your industry/domain's RCAs different or interesting?
  4. Is RCA well-integrated and supported in your industry/domain?

Here are my answers.


  1. Industry: commercial nuclear power generation.
  2. RCA Issues: Anything and everything, depending on significance or risk of a problem. (Personnel safety, equipment problems, process deficiencies, organizational issues, project / work management failures, etc.)
  3. RCA Diffs: We use a LOT of different tools, our reports can be quite long, and we put a lot of focus on risk reduction as well as recurrence prevention. Also, we have an abbreviated form of problem analysis called Apparent Cause Evaluation; it's used for problems with medium risk / significance.
  4. RCA Supported: Yes, very much so. Our regulators and industry peers demand as much. We take it pretty seriously. (We have to.)
Regards!
Bill
 
B

bpritts

Involved - Posts
ibillwilson said:
Hi All,

I approach Root Cause Analysis (RCA) in a particular way due to my industry. I'm trying to learn more about how others do RCA. So, I'm curious... what makes the way your industry does RCA different?


  1. What industry or domain is the setting for your RCA efforts?
  2. For what kinds of issues do you perform RCAs?
  3. What makes your industry/domain's RCAs different or interesting?
  4. Is RCA well-integrated and supported in your industry/domain?

Here are my answers.


  1. Industry: commercial nuclear power generation.
  2. RCA Issues: Anything and everything, depending on significance or risk of a problem. (Personnel safety, equipment problems, process deficiencies, organizational issues, project / work management failures, etc.)
  3. RCA Diffs: We use a LOT of different tools, our reports can be quite long, and we put a lot of focus on risk reduction as well as recurrence prevention. Also, we have an abbreviated form of problem analysis called Apparent Cause Evaluation; it's used for problems with medium risk / significance.
  4. RCA Supported: Yes, very much so. Our regulators and industry peers demand as much. We take it pretty seriously. (We have to.)
Regards!
Bill
Click to expand...
I am glad to hear that people in the nuclear power industry take problem-solving seriously!

I am in the business of manufacturing/distributing brake parts for heavy trucks, and we take problem solving seriously too. I drive on the same roads with these trucks!

I am not sure what we do is "special" but in the spirit of the inquiry I will mention some aspects of our methods.

** Focus on PRODUCT PERFORMANCE risks - we set our priorities based on the eventual risks of product performance, or, more specifically, the risk that the product won't perform. Many end customers just look at in spec vs. out of spec. We try to consider what the risk to actual product performance -- can we stop the truck? as our priority.

** Separate PRODUCTION root cause from FAILURE TO DETECT N/C PRODUCT root cause. These are 2 different problems. They should be considered independently. Both are important but the issues are different.

** IRREVERSIBLE CORRECTIVE ACTION. I learned this bit of jargon long ago. The concept is, that true corrective action cannot be reversed. "Re-training the operator" will fail when you get a new operator. But a change to the tool design will live on.
Some of my Chinese suppliers use the excuse "we had new operators because of turnover at Spring Festival!" I say, well, next year you will have more new operators at Spring Festival.. what happens then?"


Hope this helps, and I am interested to hear what others say.

Brad
 
B

bpritts

Involved - Posts
I realize that I missed one important point from the original question.

METHOD of RCA: In automotive, we are biased toward the 8D problem solving method. Many posts on Elsmar discuss this method, and I believe that this "bias" is well founded. That doesn't mean that we always do it well...

Ford is the role model in automotive for problem-solving/ RCA. Suggest that you consider them as a benchmark.

Brad
 
P

PenutButer

Starting to get Involved
ibillwilson said:
Hi All,

I approach Root Cause Analysis (RCA) in a particular way due to my industry. I'm trying to learn more about how others do RCA. So, I'm curious... what makes the way your industry does RCA different?


  1. What industry or domain is the setting for your RCA efforts?
  2. For what kinds of issues do you perform RCAs?
  3. What makes your industry/domain's RCAs different or interesting?
  4. Is RCA well-integrated and supported in your industry/domain?
Click to expand...

I am in Aerospace manufacturing.

We perform RCAs in conjunction with Risk Analysis or Corrective Actions. Since we are "Build to Print" facility we don't have R&D concerns to deal with. As far as the types of issues, they can span from part rejections to documentation conflicts to facility problems. The bulk of my RCA's come from the CA process though.

I like and dislike the aerospace industry RCA practices for the same reason. there are so many different methods used, each bringing their special niche to answering the question. But, it's easy for individuals and organizations alike to become set in their method and it can become a problem. Though, a fair amount of the customers and suppliers I work with, and even friends in the same field and industry, are pretty well versed in various techniques so it can be fun seeing how someone would use the 8D approach or an Ishikawa diagram where I may use some other method or combination of methods.

RCA is a core part of our NCR and corrective action processes. Most of the management and supervisory staff are fans of 'The Logical Thinking Process' by William Detmer and use a lot of the techniques from his and Goldratt's work.
 
Q

QAMTY

I have seen some companies, most of them with high risk processes, additional to 8D and fishbone use Taproot, interesting metodology to address CAs, Take a look to it.
Hope this helps
 
V

vivkrish

PenutButer said:
I am in Aerospace manufacturing.

We perform RCAs in conjunction with Risk Analysis or Corrective Actions. Since we are "Build to Print" facility we don't have R&D concerns to deal with. As far as the types of issues, they can span from part rejections to documentation conflicts to facility problems. The bulk of my RCA's come from the CA process though.

I like and dislike the aerospace industry RCA practices for the same reason. there are so many different methods used, each bringing their special niche to answering the question. But, it's easy for individuals and organizations alike to become set in their method and it can become a problem. Though, a fair amount of the customers and suppliers I work with, and even friends in the same field and industry, are pretty well versed in various techniques so it can be fun seeing how someone would use the 8D approach or an Ishikawa diagram where I may use some other method or combination of methods.

RCA is a core part of our NCR and corrective action processes. Most of the management and supervisory staff are fans of 'The Logical Thinking Process' by William Detmer and use a lot of the techniques from his and Goldratt's work.
Click to expand...
Dear,

In some of the industries RCA will be done based on 5WHY method only..

Hope this is the easiest way to get root cause.

Thanks.
 
P

PenutButer

Starting to get Involved
vivkrish said:
Dear,

In some of the industries RCA will be done based on 5WHY method only..

Hope this is the easiest way to get root cause.

Thanks.
Click to expand...
I have worked with people who live and die by the 5-Whys and I think it is excellent at quickly working through RCA. The problem I have with it is that I have seen it used as a way to prove a predetermined root cause and not necessarily to find it. I am also not a fan of the singular path concept. However, when I am working a very linear process I will more than likely fall back on it.
 
R

randomname

I'm not sure why people believe that 5-whys is a singular path tool. There can be as many paths as are necessary to find the root cause(s). Here is a simple example:

L1 - Received wrong DVD from online retailer

L2A - Individual who pulled DVD from shelf grabbed the wrong one
___L3A - Individual's eyesight has deteriorated
_____L4A - Company does not require/perform eye testing
___L3B - Lighting in the area where DVD was stored had gotten dim
_____L4B - Company does not assess lighting on a regular basis

L2B - Individual who packaged DVD did not confirm right one
___L3C - Rushed at end of shift
_____L4C - Poor resource scheduling process
 
A

AFGranda

After trying several problem solving tools during my 15 years tenure in the Aerospace industry I'd suggest DIVE - Define, Investigate, Verify, Ensure - in 1st place and 8D as 2nd choice. In both tools 5 whys and cause & effect diagrams could be utilized to uncover the true root cause.

One common pitfall of RCA is consider the direct cause as the root cause. For example when I review RCAs either from supplier or internal production sites and I see the statement: "root cause: operator error", I just reject it because operator error is actually a direct cause. Why the operator made the error? ....why, why until uncover the real root cause.
 
You must log in or register to reply here.
Top Bottom