Suggestions on Improving Containment Methodology

[Bill Ryan - 2007]

I'm looking for some ideas that I can put on the table to help my company improve our internal containment/purge procedure. Since I've started working here (18 yrs. - aaugh!!), and from what I've heard around the industry, this is our number 1 issue with letting nonconforming product reach our customers.

We have decided that the "Quality department" will write the level 2 procedure as production is at its wits end as to how to improve the situation, short of writing people up in the disciplinary process (which, of course, solves everything. ). We have a departmental meeting set up for next week and I've included some of the thoughts I want to discuss just for starters.

* All employees need to be retrained in NCM procedures

* All NCM product must be sorted by the operation which caused it (their people) even if it is found downstream

* Sorting must be done off line and cannot be done as part of the next operation, even if times allows

* In order to clear the sort the broken tools and steel must be turned in with the NCM tag as part of the sort. The supervisor signing is responsible for this (added benefit of evidence to help with CA)

* Sorts to be preformed with gauging wherever possible as the corporate goal to take the human element out of the sort (example: broken core, broken tool bit, etc.)

* Review of upstream parts

* Definitive establishment of break point (with proof)

* Review of parts or check off for parts in other areas of the shop

We have a 3 shift operation (die cast/machining/"light" assembly) with most of the newbies on the graveyard shift where supervision is also weak.

I want to thank anyone, in advance (probably later, too!) who can share their ideas or what works for them.

Bill

 

[Mike S.]

A couple quick thoughts to consider - I may have more time later.

I'd want to know why current procedures/rules are not being followed. Ask some of the apparently "guilty" why/how they allowed NCM to move ahead. Sometimes what we think is stupidity/negligence turns out to be something else, like a supervsor (or higher) verbally overruling the procedures.

I'd try to get ideas from the shop floor level people to improve buy-in and improve ideas for improvement. Doing it w/o their input & involvement is likely to fail. The more involvement the better.

If the bad parts are un-usable (not reworkable) how about something easy like red/orange spray paint sprayed on the parts as soon as the NCM is detected to make visual ID of bad stuff easy? Or easily available sticky red-tags, dye, etc.

This may take on-site babysitting for awhile -- living and working with the shop people for awhile to see what is going on. Sometimes much good can come of this even on issues beyond the initial problem.

 

[D.Scott]

I think I am seeing a sceanerio where you don't have a problem identifying when a containment is needed. On the other hand I am thinking if you get nonconforming product product to the customer this identification isn't all that effective. The fixing of a sort procedure is fairly simple - find where it is broken and fix it.

I would almost suspect that as this is a major concern, the "need to confine" is not always recognized. Alternatively, when it is recognized, it is ignored for some reason.

We determined that certain situations would cause us to be suspicious of potential problems so we decided to develop our containment around a sample inspection whenever one of them was present. Our list follows:

1.0 Whenever a gross nonconformance is discovered in a container of parts, that container must be 100% sorted to ensure no other gross nonconformances are present.

2.0 Whenever set-up parts are being run, all the parts run during that set-up must be isolated and 100% sorted to ensure no nonconforming parts are present.

3.0 Whenever an operator suspects there is a possibility that gross nonconformances may be in the container of parts they are running, those parts must be isolated and a 2000 piece random sample must be thoroughly inspected. If one or more gross nonconformances are found, the entire container must be 100% sorted to ensure no other gross nonconformances are present.

4.0 Whenever excessive running problems occur in production, the container run must have a 2000 piece random sample inspection. If one or more gross nonconformances are found, the entire container must be 100% inspected to ensure no other gross nonconformances are present.

5.0 Whenever a trainee has been assigned to run a production part, the container run must have a 2000 piece random sample inspection. If one or more gross nonconformances are found, the entire container must be 100% inspected to ensure no other gross nonconformances are present.

6.0 Whenever a new part is being processed or an inactive part has been reactivated, the first three lots must have a 2000 piece random sample inspection. If one or more gross nonconformances are found, the entire container must be 100% inspected to ensure no other gross nonconformances are present. The new part would then be subject to three additional clear inspections on future production runs.

7.0 Whenever a customer concern (CCR) is received, that part is subject to 100% inspection to ensure no nonconformance is present. Determination of the need for and frequency of this inspection will be made by corporate quality and specified on the CCR.

8.0 Whenever new or refurbished tooling, which would effect the end product results, is run for the first time, the container run must have a 2000 piece random sample inspection. If one or more gross nonconformances are found, the entire container must be 100% inspected to ensure no other gross nonconformances are present.

9.0 Whenever there is a process or material change, the container run must have a 2000 piece random sample inspection. If one or more gross nonconformances are found, the entire container must be 100% inspected to ensure no other gross nonconformances are present.

10.0 Whenever there is a test/inspection failure according to defined acceptance criteria, the product must be placed on hold pending disposition by Quality. In the event the product has been released under customer waiver.

Obviously these are not the only things that would require a containment but they are the largest contributors for us. You also may find another problem if your production department has goals/quotas. Containment slows down the process so a potential problem exists in employees who are afraid to report a containment situation.

Dave

 

[SteelMaiden]

Bill,

OK, it sounds like you've identified the cause as operators ignoring the need for containment, or at least that is how I read it.

a) You are probably on the right track to "start over" so to speak and review current procedures and rewrite them.
b) Training is good, but make sure that you give them good info on what is acceptable and unacceptable
c) Making the "culprits" take care of sorting has worked for me before
d) buy-in, motivation, whatever you want to call it appears to not be real high on the list, eh?
Include the production employees in information as to how much money the company looses when nonconforming material is shipped out. Make sure you post (share) this monetary information for all areas, though, or it will appear that you are targeting only one group. If data entry errors caused material problems, put those figures down, along with each production line etc. Make sure you publicize this widely, along with some goals and plans for improvement. Share the information often enough so people see the changes. If you don't work under a bonus system now (if you do, you might see if you can get management to take a hit out of the offending shift/crew's bonus) maybe you can talk management into catering a lunch for the facility if your goals are met within some defined time frame.

Be tough when you need to, but make sure that you also reward good "behavior". There is nothing worse in the eyes of a production employee to get reamed for stuff, but never hear an "Atta Boy" when you change your attitude to improve profitability/effeciency.

Of course, what you can actually do might depend on how your business is run, union or nonunion, etc.

 

[Bill Ryan - 2007]

Thanks gang!!

I didn't mean to imply that our dept. was going to write the procedure and ramrod it through to the production people. We will, probably, write the Draft as we'd like to see it and then invite the production supers and operators to revise and come up with the final version.

Mike - One of the problems on our floor is that we (QI) will sign off on something that will not affect the customer with regards to fit/function, yet want the issue corrected on the tooling. After an operator has that happen a few times, he begins to think he "knows" what our response will be and just bypasses the NCM ticket. Another thing you mentioned was their super might have told them "not to worry about it" - that has happened a lot in the past but I think we have impressed upon them that we need to be in the loop.

Dave - We are, actually, fairly good at identifying a nonconformance and even sorting that container (now that we have retrained our die cast operators that gages are not hammers and require a little care ).

SteelMaiden - I think you are correct in that my company still has too much of a punitive mindset. When operators "get in trouble" for questioning whether something is OK or not, it's hard to blame them for overlooking "marginal" types of issues. There aren't a whole of "Atta boys" around here yet and when there are, they seem to be confined to the management team - not trickled down to the floor (please don't think it's like a concentration camp here, there are some rewards for great production months - such as free lunches - and the Supers do have monies budgeted for gift certificates for "high performance" rewards - which I'm not sure I'm in favor of)) . We are non-union so there are actually times where management and "hourlies" truly do work together, but not nearly enough instances.

Our biggest issue is that once we identify an issue and it is determined that it must be corrected and not shipped, we don't go back far enough (e.g. - to the shipping dock) to make sure we have our arms around ALL suspect material. It seems to be much easier to think we got the "start point" in the container where the issue was found. There isn't a lot of attention paid to, for instance, cavity number or whether the issue is "intermittent" or not. I don't feel like I'm expressing what I want to properly but basically, I'm looking for some way to be sure we go back far enough to make sure we have captured the "start point" and then purged all inventory as needed.

Bill

 

[PA-Engineer]

Re: Seeking Suggestions on Improving Containment Methodology

Hello
I appears it has been some time since this thread was active but I wanted to give the topic a little steam. I am also looking for ways to improve our containment actions. We have been able to identify WHEN to contain but seem to lack the systems/procedure to verify that the procedures have been followed properly. I am looking to hear ideas about a robust method (and simple method) for our operators to contain parts after a broken tool without affecting production. We have kicked around ideas which have included a "floating" person to monitor tooling usage and containment procedures but there has to be a better way.

Any help/ideas would be greatly aprpeciated.
PA-Engineer

 

[CarolX]

Re: Seeking Suggestions on Improving Containment Methodology

Well, I can't help too much here - but I am glad to see this resurected - as this is a big issue for us right now. I just had this issue added to our CI program. As the discussion was underway in our CI meeting - I saw my engineer's eyes roll to the back of his skull when he began to understand the issues.

Right now - I am preparing an overview for next weeks meeting - and I am looking for a good definition for containment that my co-workers will understand.