A
Following what Pancho and Qcace have said - you can't determine the root cause up front.
The reporter can only be sure about the symptoms, so just report the symptoms.
for example:
I can't do my job because something is wrong:
Bad parts - wrong size, wrong material, no material, ...
Bad information - order missing critical information (cut length, colour, quantity, steel grade, ...)
Bad process - tools out of calibration, machine not capable of +/-0.x
Bad planning - can't meet a delivery date of yesterday
The list you mentioned earlier is probably right.
And your gut feeling - about bad communication - might be right as well. That's the conclusion I came to with the data i collected.
If you can figure out a way of telling that to management without them getting personally offended, let me know. (I think Demming had a way)
The reporter can only be sure about the symptoms, so just report the symptoms.
for example:
I can't do my job because something is wrong:
Bad parts - wrong size, wrong material, no material, ...
Bad information - order missing critical information (cut length, colour, quantity, steel grade, ...)
Bad process - tools out of calibration, machine not capable of +/-0.x
Bad planning - can't meet a delivery date of yesterday
The list you mentioned earlier is probably right.
And your gut feeling - about bad communication - might be right as well. That's the conclusion I came to with the data i collected.
If you can figure out a way of telling that to management without them getting personally offended, let me know. (I think Demming had a way)
Thank you for the reply pancho.
I actually do have a root cause area but this is derived during the investigation.
My main form within my database where investigators of a nonconformance work is broken down in to the following pages:
nonconformance overview = all information submitted initially by the reporter
costs = allowing multiple costs to be assigned to the nonconformance as they are realised, this is broken down to be able to add in estimated/actual and assign them a category and department who has incurred them.
Investigation = This allows for notes of who has been involved, methods used and then an area to report each root cause (as there can be multiple). Root causes are also categorised for example, not following procedure, out of date documentation etc. A start date for the investigation and an end date. Finally the outcome of the investigation which for now is two options of "Use as is" or "correction required". This has a reasoning box next to it so they can describe why the conclusion has been reached and who has signed it off.
Corrections = Corrective actions which are broken down in to seperate actions/start dates/proposed end date/actual end date who is the owner/who is carrying the action out and what the action is. Then finally a section to mark the nonconformance for closure and recommendations.
So as you can see, there is categorisation right through the sytem which is great for analysis however, I do not want to have to use qualitative analysis to understand what is actually being reported initialy. I would prefer the initial reporter to suggest what category their nonconformance event description falls in to. This would then allow me to do trend analysis on a department and business perspective associating costs and root causes along the way. I can then help to steer the business in the right direction not just the departments.
But as we have both said, it is hard to make choices on what the initial categorisation is without driving down in to a root cause. By all means I do not mind users stating what they "think" but I have always said you should let the data tell the story and not be driven by "gut feelings". However, with that being said if I can categorise nonconformance events coming in, I can use data to drive priorities and do the right things first.
To put it in perspective, our business is looking at 1,300 internal nonconformances this year, amounting to profit loss in the millions (wasted time, rework hours mainly)
. We have chosen to use the categorisation of events to show us what our biggest issues are (pareto analysis easiest way) and then do thorough investigations on these and bring their occurence rate down thus reducing costs until we are at a manageable level to investigate everything coming in to the system thoroughly.
p.s before anyone says it, we do actually have a nonconformance system which is concerned with customer delivery however my system will be lookin at internal nonconformance, at the same time the first system is becoming very out dated and support for it is set to end soon so our quality department are looking for a replacement which may end up being this one once it has been fully tested/refined/audited.
I actually do have a root cause area but this is derived during the investigation.
My main form within my database where investigators of a nonconformance work is broken down in to the following pages:
nonconformance overview = all information submitted initially by the reporter
costs = allowing multiple costs to be assigned to the nonconformance as they are realised, this is broken down to be able to add in estimated/actual and assign them a category and department who has incurred them.
Investigation = This allows for notes of who has been involved, methods used and then an area to report each root cause (as there can be multiple). Root causes are also categorised for example, not following procedure, out of date documentation etc. A start date for the investigation and an end date. Finally the outcome of the investigation which for now is two options of "Use as is" or "correction required". This has a reasoning box next to it so they can describe why the conclusion has been reached and who has signed it off.
Corrections = Corrective actions which are broken down in to seperate actions/start dates/proposed end date/actual end date who is the owner/who is carrying the action out and what the action is. Then finally a section to mark the nonconformance for closure and recommendations.
So as you can see, there is categorisation right through the sytem which is great for analysis however, I do not want to have to use qualitative analysis to understand what is actually being reported initialy. I would prefer the initial reporter to suggest what category their nonconformance event description falls in to. This would then allow me to do trend analysis on a department and business perspective associating costs and root causes along the way. I can then help to steer the business in the right direction not just the departments.
But as we have both said, it is hard to make choices on what the initial categorisation is without driving down in to a root cause. By all means I do not mind users stating what they "think" but I have always said you should let the data tell the story and not be driven by "gut feelings". However, with that being said if I can categorise nonconformance events coming in, I can use data to drive priorities and do the right things first.
To put it in perspective, our business is looking at 1,300 internal nonconformances this year, amounting to profit loss in the millions (wasted time, rework hours mainly)
. We have chosen to use the categorisation of events to show us what our biggest issues are (pareto analysis easiest way) and then do thorough investigations on these and bring their occurence rate down thus reducing costs until we are at a manageable level to investigate everything coming in to the system thoroughly.p.s before anyone says it, we do actually have a nonconformance system which is concerned with customer delivery however my system will be lookin at internal nonconformance, at the same time the first system is becoming very out dated and support for it is set to end soon so our quality department are looking for a replacement which may end up being this one once it has been fully tested/refined/audited.