Printer Friendly

CIM is part of the big picture at Kodak.

The "Catches of the Week" at Eastman Kodak Co.'s Elm Grove Plant in Rochester, N.Y., aren't fish from nearby Lake Ontario and they don't appear on the cafeteria menu. These "catches" are plugged water lines, leaky injection nozzles. broken heater bands, and the like. You'll find them listed in Kodak's biweekly Process Monitoring Center Newsletter as the culprits behind processing irregularities "caught" by the CIM team in the company's Precision Plastics Organization (PPO).

PPO produces all the plastic parts for Kodak's line of single-use cameras and also has a small but growing custom molding business. In the Process Monitoring Center, a room filled with 14 computer screens, PPO's CIM team monitors every heartbeat of 92 injection presses. They look for telltale warnings so they can "catch" problems before they happen. They also try to trim the fat off cycle times. And they continually evaluate the ability of machines and molds to make a good product.

But CIM isn't just the responsibility of the CIM team at PPO - it's everybody's business. CIM is central to the training in "6-sigma quality" and "continuous improvement" required of everyone in PPO. CIM has produced astonishing productivity increases, sufficient to enable PPO to scale back from seven-day operation to five days (though the group is now back to working seven days to satisfy swelling demand for its product).

Computer integration is a cornerstone of PPO manager George Fellows' push for "worldclass" quality and efficiency. CIM has moved the operation's quality focus from parts inspection to maintaining a uniform process. And it has been instrumental in achieving and maintaining ISO 9002 quality certification (awarded in 1993).

For all these reasons, the newsletter of the Process Monitoring Center has become popular reading material for all 435 PPO employees.


In June 1991, the first 16 of PPO's molding machines were networked to a CIM system from Hunkar Laboratories Inc. of Cincinnati. Today, 92 machines from 28 to 1000 tons are monitored from three Hunkar CIM stations and several standalone PCs in the Process Monitoring Center. A fourth CIM station in another location monitors five more presses that constitute a specialized cell for medical custom molding.

Since the first installation of CIM, PPO has achieved productivity gains equivalent to the capacity of 14 additional injection machines. A big part of that gain came from optimizing cycle times with the aid of process monitoring. Kodak monitors 14 variables on every shot. "CIM helps you to better understand process variables so you can manipulate them," says Mike Lee, head of the CIM team and the Methods Engineering group responsible for process helped Methods achieve an average 30% reduction in cycle times, he reports. For PPO's oldest machines, some of which had seen more than 20 years' service, CIM sliced a whopping 42% off average cycle times. CIM also helped identify the presses most ripe for retirement.

Lee emphasizes that these improvements were achieved with existing tooling. But CIM data has helped spur development of new product- and tool-design concepts that will make PPO even more efficient, he says (see p. 92).

Another boost to productivity came from scrap reductions. "Before CIM, we might go a whole day making bad parts on a press before we discovered that a water line had been turned off," Lee recalls. But the PC screens in the Process Monitoring Center immediately warn of a process deviation and give clues of where to look for the cause.

One of the greatest values of real-time process monitoring is that it helps pinpoint machine maintenance needs even before any bad parts are produced. Lee cites the example of check rings: "As soon as one shows any variation, it's out of there."

But when there is a problem, the process monitors pinpoint the bad cycles. "We can tell exactly where to find the bad parts. Before, you'd have to sort through a million of them," Lee says.


The Methods group is constantly exploring creative ways to make use of the mountains of data the CIM system accumulates on disk. The wall behind Mike Lee's desk in the Process Monitoring Center is papered with dozens of charts displaying CIM data in different ways. One bar chart ranks every one of the 97 presses according to its ability to maintain [+ or -]6-sigma control limits that are at least as good as the average for all PPO's machines. Another chart shows the growing number of machine-and-mold combinations that have been analyzed for statistical process capability - more than 1800 to date. There are also charts of Catches of the Week, showing which sources of problems (water, nozzles, tooling, etc.) crop up most frequently. Still more charts show which machines require the most maintenance and perhaps ought to be replaced.

"Process windows" are the most important and probably most innovative use PPO has made of CIM data. Process or moldability windows are now determined for every new tool that comes into the shop, explains Mark Trudeau, a statistical process engineer who played a central role in CIM implementation. In a one-hour Design of Experiments (DOE) procedure, the window is determined by examining part quality while deliberately varying selected parameters - usually boost and hold pressures. Statistical analysis of the CIM data also defines the 6-sigma limits - that is, the range of variance within which 99.9996% of all cycles are expected to fall.

Figure 1 shows a process window in which the 6-sigma limits are constrained within a region that is well centered within the zone that yields acceptable parts. Figure 2, by contrast, shows inability to hold as tight 6-sigma limits. What's worse, the 6-sigma region overlaps the zone in which flashing occurs. Either the press or tool must be reworked to provide greater confidence in molding good parts.

"The process window lets you see crazy things," Lee points out. In Fig. 1, for example, flashing can be eliminated by raising boost pressure. "This is unusual - eliminating flash by raising pressure!"

Now that sources of machine variation have been reduced, Lee is starting to use process monitoring to detect resin variability. "When different shipments of the same resin behaved differently, the supplier said we had a process or moisture problem," Lee recalls. "But we were able to demonstrate our process consistency and identify the differently performing resins as coming from different plants of the same supplier. We now get resin only from the one plant whose product worked better for us."


Commitment to continuous quality improvement is part of the new philosophy at PPO. To a large extent, CIM has made that change possible. By focusing on 6-sigma quality, PPO strives on every job to hold to the smallest molding window the mold and machine will allow - even if that goes beyond the customer's required tolerances. "We may eliminate a problem the customer didn't even ask us to fix," Lee says.

One way CIM has made this change possible is that it "adds the ability to relate the product to the process," according to George Fellows. "Until we focused on process stability instead of product stability, we had no idea of what world-class really meant."

"CIM has taken quality and raised it to a new level," says quality-assurance coordinator Bill Russell. "We're backing away from product inspection and going toward process monitoring. SPC has eliminated a lot of inspection. It pushes the q-c process down to the operators." PPO is installing new operator terminals that display all quality data at the press.


All new operators with less than a year's experience are trained in the Process Monitoring Center. With the CIM data, Lee says, "they get more satisfactory answers to why a problem occurred than even the most experienced operator can give them. The learning curve is much shorter in here." He adds, "I have 24 years' experience, but I've learned more in the past two years with CIM than at any time in my career."

In truth, CIM has become part of the education of everyone at PPO. Every employee is required to take a minimum of 90 minutes' instruction in "Six Sigma Awareness," which consists of 20 topics in basic SPC and process-monitoring principles.

The process-monitoring newsletter helps continue that education. One regular feature is a test of readers' abilities to interpret an actual process window. On the process window in Fig. 2, for example, readers might be asked, "Hold pressure has no effect on whether the part is short. True or false?"
COPYRIGHT 1995 Gardner Publications, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1995, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Plastics Technology's Annual CIM Leaders Award; computer-integrated manufacturing; Eastman Kodak Co.
Author:Naitove, Matthew H.
Publication:Plastics Technology
Date:Nov 1, 1995
Previous Article:Plastics Technology's PC software buyer's guide.
Next Article:Ten years after: CIM leaders revisited.

Related Articles
Molder's growth extends CIM's reach.
Ten years after: CIM leaders revisited.
At Gillette Co.: monitoring automation keeps scrap at bay.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters