MTAC Workgroup #132 Six Sigma IMB Best Practices
MINUTES
October 9, 2009, 2009, 10:00 AM
Susan Pinter: Did everyone get a chance to read the minutes; I know it was kind of last minute. I sent them out yesterday along with the report that Tim had created. Did anyone have any questions or concerns with them?
Anita: I did have a question on the report. On the best practices report on the machine set up; number 11, it said just Kodak printer, 22 pitch forms, 120 BPI adjusting the feed rate. Not everybody has Kodak equipments, so I didn’t know if there any plans to put in there maybe the top 3 most popular ink jet equipments that we’ve come across; or are we just going to leave it as just one spec.
Jesse: Which document was that again?
Anita: Page 14 of the continuous improvement document under machine set-up, item number 11.
Ernest Rodriguez: This one is specifically for the Kodak. There is others that are for the Domino?
Susan: Ernest, will there be a best practice overall for other equipment since this one is geared just towards the Kodak?
Tim O’Reilly: Obviously we can only address the types, manufacturers, print heads that we were privy to during our review. We certainly did see a couple more than the Kodak, which we could address, but we cannot address everything that’s out there.
Anita: No, certainly. I don’t think there’s a huge variety, though. I think most people have tended to stick to 3 or 4 main types.
Susan: Tim, we have you set up to talk about the plant reviews. Then go into your report out.
Tim: I’ll just start from the beginning where we’ve completed 3 tiers, 3 site reviews. We have another tier 1 site review scheduled for next week at a letter shop. That will be the last visit that we’ll be doing. We’ve also completed 3 tier 2 businesses. They were all done last week. What our plan is to do is by probably Tuesday the 20th or maybe even the 21st, we plan on having the final document available and send it out to Susan so she can decimate it to the rest of the group. We would ask by the 27th that any comments or concerns be sent back to us so that we can review them, and as necessary incorporate them back into the document. By the 29th, we will issue the final. We did, from the last meeting, we had a couple weeks to go—we took companies’ names off of the documents. There’s no reference to them whatsoever. While we were out there, we did look at 3 different types of ink as well as the printers. We got to see the acetone, the water, and the erm. As discussed earlier, we saw numerous print heads. The
equipment that we looked at ranged in age from just a few years old to 16 and even 30 years old. What we did find is that there is a template the Postal Service makes available for looking at bar codes. We found old templates out there. There’s a date on the template that’s the most current. We found ‘07’s out there. That certainly is best practice too, to make sure we have the most current, up-to-date templates. We found in-line verification. There’s also an I loop available the can be utilized to try and diagnose the bar codes. That was not, for the most part, being utilized either. It’s probably a best as well to use the eye loops. Companies in tier 2 were using a bar code reader. Bar code reader was accepting the bar code, but it really isn’t dynamic enough to be used with the IMB. So we weren’t good results on their bar code reader when we would run it, we would find that the acceptance rates were low. We did use this document to help improve tier 2’s. We were able to do that in every case. Actually greater than 95%. Most of our focus was on the table where the IMB and address is printed. Maybe 4 foot long or so. There is some issues that we observed; the chains, the drive, the table needed to be typed. That was actually a very important step. They were out of tolerance, as I think the one maintenance person referred to a lot of slop. That caused some variation and movement in the piece, which then caused issues for the IMB. Make sure the print head was clean. We also ran into a circumstance where the guide rails, if that’s the term you guys use—what keeps the piece going through that table. In one case, those guides were too tight. The piece was being forced, or pushed, through. That caused some issues for the IMB as well. We looked at poly.
It is a concern. We ran it through the ERM 3 and after adjustments we made, we were able to in the adjustments move the IMB to the top of the address. We also had to tilt the print head. They were using a laser, ink jet on that one. We had to twist it ever so slightly to get rid of the pitch there. That was on with the white stripe on it.
Susan: Tim, can you explain what the ERM is?
Tim: The actual name is Envelope Reflecting Meter. It’s a machine that – it’s not too large. It’s probably 1½ foot x 8 inches deep and high. It has a camera in it at the far end, and it has a window that you’re able to put the piece of mail into the window. The piece is held on with pressure. It’s a spring-loaded round device that holds the piece up against the window. You have software that the ERM3 is connected to a lap top. You have software on the lap top. You’re able to take a shot of that piece and then run it through the software and it comes back and tells you type of errors, were there failures, were there warnings. One thing the ERM doesn’t do is in Merlin if, I’m comparing it to Merlin. In Merlin if you have more than 10 warnings, it’s a failure. That doesn’t come out that way with the ERM. You count the warnings and say to yourself OK it’s below 10 so it’s just warnings, or its 10 or above it’s a failure. It gives you a real good comparison as to what Merlin’s going to do in most cases. I’m going to talk to you about the issue poly going thru the Merlin here in a few minutes. Other than poly we didn’t find significant differences between what ERM read and gave us issues and concerns on than what Merlin did. We were able to in the poly case use the ERM, we got it up to about 95% accept rate. We also test it and put on these large white labels on the poly. Believe it or not there was no difference in having the right label on the poly or just using the poly itself when we read it through Merlin and through the ERM. The Merlin accept rates were very low (about 12%), probably because of the wrinkled plastic. Merlin feeds it right through and it’s more or less a continuous feed where ERM we were doing one piece at a time, and it was underneath pressure, so the wrinkles were taken out. We’ve had some discussions on how we can try to improve Merlin with this. There will be more discussions with our engineering group. As we were pushing it through Merlin, maybe we could use a burst of air pressure to flatten it out, then try to get a good reading. Obviously we’ll get the engineers involved and see what they can do. We also had asked if the sheet cutter, if that equipment on the bagger could be adjusted so that if there’s less “waste” of the poly on each side, that would tighten up the bag and you’d have less variation. That we were told was not possible. The equipment required an inch between each of the catalogs. We’re probably going to do a little bit more further testing. Maybe taping down that extra poly, and running it through again and see if that improves the rate. We have intentions of having a meeting with the manufacturers of some of this equipment, and have some discussion about what they could do from their side so that there could be improvements in this. Maybe they can adjust some of these sheet cutters. Maybe they can design machines where there are chains that become loose and floppy. There could be self-adjusters on them so that there’s no need for the operator to get under there and make the adjustments. Because in the one case that we saw, it took maybe 15-20 minutes for an operator to get under there and adjust the change. When he came out, he was greasy up to his elbows. Maybe there are some things the manufacturers can do that will help for the future. Obviously they can’t help us right now, but maybe we can get some help for the future. When the first run for the IMB was put on the poly, it was at the bottom of the address window. We asked if it could be put up at the top because there was a seam up closer to the top. That maybe we get more stability if we have the IMB up on the top. That absolutely worked, and that’s why the thought was if we were able to tighten down that excess waste on either side of that catalog, by you having the sheet cut it closer that it also may improve the readability for us. I think we also have increased their font size. I talked earlier they twisted the head slightly to get rid of the pitch errors.
Susan: When they were ink jetting on the poly, were that directly on the poly or was that on the label.
Tim: Most of that was directly on the poly. There were some test pieces that we have to put a label on, and then do the ink jet over it. Again, we had no differences between ERM and reading those with labels on vs. directly onto the poly.
Charlie: How about if the address was inside underneath the poly.
Tim: We did not get to observe that. That was not part of the production where we were visiting.
Susan: Charlie, do you think that will still be an issue if the book if ink jetted and then the pieces put in poly. Because we’re not running the poly pieces on Merlin right now anyways.
Tim: I guess the bottom line is that we were able to go into the sight using the document and also to be honest with you, we did learn more while we were there. We were able to improve every sight that we visited, at or above the 95%. The plant that I visited, the first day that we were there, adjustments were made to one line and they were very successful. So we went back 2 days later, the final day, and we re-tested that line and those adjustments were still in place. They were still in the mid-90 for acceptance rate, after 2 days production. The place were at is run 24-5, so quite a few pieces run through that piece of equipment over the course of 2 days. Those adjustments did stick, and they did work. As we were leaving, wrapping up the ERM3, one of the operators came up and made a comment that it was good that we as experts could come in and help them. As I said Monday, when I was talking about this, that put a smile on my face. To be quite honest with everybody, we are not experts in IMB. I jokingly say 2 months ago, we didn’t know how to spell it. We were able to come in, using the main 6 sigma tools, and the best practices documents, and really a bunch of logic and common sense. We were able to help the folks that we had visited. We feel that we’ve been successful. The issue that we had, and Susan’s well aware of it , is that we cannot find a tier 3 location that we can visit. Basically, every option has been exhausted. We, on the clinical side, even attempted to find a location this. We were also unsuccessful. We were unsuccessful because we did get some names. Susan provided names this week and we made an attempt to do it and what our plan was they we would…We did this in 2 locations. We would ask the baler to have brought a test back for us. We made arrangements for the local plants, or customer service office at the Merlin to pick up that test and run it through Merlin and give us the results. In both cases, there were 3 different test specs runs because the one printer did letters. He also did slack. He did about 96% on one and 149% on the other. The other company did stack; they did a test spec of 100 for us. In all 3 cases, they passed with flying colors. 200 and one 96% passed. We were unable to identify a tier 3 sight that was a viable candidate, so the time has run out.
Susan: That’s the biggest concern, because the reviews that you have done have all been sights where they were all flats. Correct?
Tim: Yes:
Anita; so we have never had a review process of the letter mail. Then, especially in a 3 tier. I know next week you’re going to a tier one plant in Lansing, Michigan.
Tim: Correct. Tier 1 for letters. As you well know, we visited the sights that were provided by the industry, we’re volunteers As I told you all when we first kicked this off, if you’re getting enough sites, I could do 5 a week, and make sure they got done, we could meet our time frame. The bottom line, that neither you nor I could have done, to change who the volunteers were.
Susan: It’s been a little bit of a challenge with the volunteers. With the time constraints and this had to be done so quickly. This being fall season, people are busy. But I am a little concerned that we won’t be able to do a tier 3. Just to show what we’ve learned in tier 1 and tier 2 is going to…going into a plant that completely is not doing anything. I understand the reason why we can’t do it is because we don’t have a company. We’ve asked at different events, and we’ve been batting out on trying to get a volunteer.
Charlie: Did you go on MSF site to see if you could find somebody? They are more likely to have somebody in that division.
Susan: I know. There were different emails that went out to the different M Tech people. It was brought up at Post Com and Don and I have really been struggling trying to locate some one. The companies that we have found; again, time constraints. The company’s going to give up a couple of days for some employees to work with the Postal Service. You and both know that this time of year is pretty challenging.
Charlie: We have to remember too, though, that the couple sites that did appear viable that did volunteer. Then we tested their mail; they were companies that have never produced a mailing or they had done test pieces before. Both of those companies came in with flying colors.
Susan: That one just kind of baffles me. The one company you were talking about has never run it; and they ran 100%. We laughed about it yesterday too.