Always one root cause?

[qualprod]

Hello everybody

In my company as normal actions when a corrective action is solved, we always do the next.
we start by using the fishbone, detecting possible root causes, after going deeper, we determined that
causes of the problem were two causes.
In our procedure we include two root causes found and based on this, defined action plans.
From this assumption, I have to two doubts.
1-Could exist several root causes, or must be only one?
2-I know the actions have to be focused to eliminate root causes but....
Could I include other actions which it seems are not directly related but is done
to reinforce the efforts to eliminate any cause?
Example of problem:
Frequent bad inspection of product, which cause rejects.
needs a corrective action.

Possible root causes
-Competency, failed instruments, room for inspection with low intensity light, over inspection and limited time, working in two shifts.
-After more analysis, were determined were only
two root causes: working in two shifts and competency.
-Action plans: to give full training and working only one shift.
Could I also add an additional plan which could reinforce the performance of inspections
although are not related to root causes?
for instance to improve the room conditions, light clean, 5s's applied,etc.

Please share your thoughts

Thanks

 

[Al Rosen]

Why is working in two shifts a root cause? Why is low light over inspection a possible root cause? Brighter light would only improve detection and not eliminate the problem.

Bad inspections do not cause rejects unless the inspection process introduces defects. Building product incorrectly cause defects.

 

[L.Soad]

Why is working in two shifts a root cause?

Maybe it is not a well defined root cause, maybe it is a lack of communication between shifts. Just my thought.

 

[Ed Panek]

You need to address at least one root cause although there may be more. Lets imagine you have one true root cause but your process drives you to address two (one true and one not true). It's not ideal but it's likely to be a continuous improvement.

 

[Bev D]

There is so much to say about this.

First yes you can have more than one root cause. In fact I've started using the term causal mechanism as it is less singular. Causal systems are never flat and rarely linear with only main effects. Interactions are all too common where there are at least two factors that must interact. Often there are conditions for failure that interact with some factor or defect that results in a failure. Certainly there is room for more discipline by adding actions that will create a more conducive environment in general to reduce defects (5S, better visual aids, standard work)

We also have to remember that there is the physical cause of the defect or failure, there is the escape cause to be concerned with and sometimes we are concerned about the system failure that allowed those two causes to occur.

Next: I have to echo what others have said. inspection doesn't cause defects. it may allow them to escape through non-detection. From what you've given us I can't see that you have identified any true causes... I would want to hear a better problem statement. Is this a visual inspection or some measurement process? is it 100% or a sample size? How did you prove your causes? what data did you have? I really don't see that two shifts can cause anything. Also what do you mean by competency?

 

[Ed Panek]

There is so much to say about this.

First yes you can have more than one root cause. In fact I've started using the term causal mechanism as it is less singular. Causal systems are never flat and rarely linear with only main effects. Interactions are all too common where there are at least two factors that must interact. Often there are conditions for failure that interact with some factor or defect that results in a failure. Certainly there is room for more discipline by adding actions that will create a more conducive environment in general to reduce defects (5S, better visual aids, standard work)

We also have to remember that there is the physical cause of the defect or failure, there is the escape cause to be concerned with and sometimes we are concerned about the system failure that allowed those two causes to occur.

Next: I have to echo what others have said. inspection doesn't cause defects. it may allow them to escape through non-detection. From what you've given us I can't see that you have identified any true causes... I would want to hear a better problem statement. Is this a visual inspection or some measurement process? is it 100% or a sample size? How did you prove your causes? what data did you have? I really don't see that two shifts can cause anything. Also what do you mean by competency?

I took a class including root cause analysis. The instructor took us through a root cause analysis of the dinosaur extinction event. Was it the location of the asteroid impact? The speed? Was our atmosphere not thick enough to slow down the asteroid or incinerate it enough? too little water coverage? Too much? Was the Sun inadequate in steering the asteroid away or was it the proximate cause? What about the gravity well of Jupiter? Where did the asteroid come from? Was it loose matter from the Big Bang. Was the root cause dinosaurs ate plants that require sunlight? We determined we could address all of these things to prevent another extinction event from asteroid impacts.

Without a better problem statement its tough to help you.

 

[Sidney Vianna]

The instructor took us through a root cause analysis of the dinosaur extinction event.

That's an investigation of an event; not a RCA. The meteor stroke that led to the demise of dinosaurs was not a "SYSTEM FAILURE". A RCA focus is to understand what system breakdown(s) allowed a problem to exist.

 

[Miner]

There is so much to say about this.

First yes you can have more than one root cause. In fact I've started using the term causal mechanism as it is less singular. Causal systems are never flat and rarely linear with only main effects. Interactions are all too common where there are at least two factors that must interact. Often there are conditions for failure that interact with some factor or defect that results in a failure. Certainly there is room for more discipline by adding actions that will create a more conducive environment in general to reduce defects (5S, better visual aids, standard work)

I agree. I use the term causal system. Bev, I think I recall you mentioning the ThinkReliability approach once. There are a number of examples on their website that illustrate the approach. You can have multiple causes and corrective action is viewed as defense in depth.

Note: I am not affiliated with the company, but recommend the underlying thought process.

 

[Marc]

I used to give an in-house Problem Solving course some years ago. Typically we would take internal nonconformances and 're-investigate' them. We would often come up with 3 or 4 'root causes'. Then we would try to drill down to a 'true' root cause. Many times no definite single 'root cause' could be identified.

See: Effective Problem Solving - Six Problem Solving Fundamentals

Index of /Cove_Premium (look for Problem Solving and 8-D files).

and

Index of /Cove_Members

and https://elsmar.com/Cove_Members/8-D Problem Solving Guide Rev G3.pptx

 

[Ed Panek]

"Struck" not "Stroke" or sometimes I feel I might have one .