At Gillette Co.: monitoring automation keeps scrap at bay.
Take a Wide-Angle Look At Your Process
This year's CIM Leaders Award goes to two companies that have recognized the value of extending their shop-floor computer networks beyond primary machinery to the previously unobserved realm of auxiliary and downstream equipment.
No matter how well plantwide monitoring systems watch your primary process, they can paint a woefully sparse picture of your overall manufacturing operation. Why? Because these systems seldom encompass auxiliary and downstream equipment or plant utilities, creating what Syscon-PlantStar president Steve Thomas calls a "black hole" of important processing information. Nick Hunkar, v.p. of Hunkar Laboratories, agrees: "Real-time monitoring of auxiliaries and secondary processes can tell you a lot about your operation that you wouldn't notice otherwise."
Both CIM suppliers explain that mysterious process interruptions may be rooted ultimately in low-visibility factors such as tower water temperature, plant humidity, and auxiliary-equipment functions. The same holds true for downstream processes like decorating, finishing, and assembly. Pulling a good part off the molding machine doesn't mean much if it must make its way through maze of secondary operations, any of which can turn it into a reject. Overlooking problems in the downstream realm can result in hidden scrap and downtime. Hunkar also notes that strategic planning decisions - such as capacity planning, resource allocation, and scheduling - are thrown off when you miss the effect of downstream processes on total manufacturing productivity.
In the past, the ability to wire auxiliary and downstream equipment into a plantwide monitoring network was hindered by a lack of communication capabilities in non-primary equipment. During the past few years, this barrier has started to fall away. More and more types of auxiliary equipment have communications capabilities today, in part because of industry standards like the SPI protocol.
As communications capabilities of auxiliary and downstream equipment improve, the possibilities for monitoring grow increasingly abundant. Some clever processors today monitor a full range of auxiliary equipment - including mold-cooling systems, materials handling, dryers, air compressors, silos, environmental controls, incoming power voltage, and much more. Downstream monitoring currently includes parts removal, automated packaging, finishing, and assembly.
When you mold upwards of 15 billion parts per year, productivity yardsticks like scrap rate and uptime loom large. To drive waste down to a minimum while keeping its manufacturing cells running smoothly, Gillette Co. has adapted an innovative CIM system that reaches beyond the 105 molding machines in its Boston operation to embrace the company's automated parts-removal and handling systems.
The company has also taken steps to ensure that its gargantuan data stream is useful to everyone from machine troubleshooters on the shop floor all the way up to decision makers in the executive suite. "Information has many different customers," says operations manager Mike Estrada, noting that Gillette developed its own customized production-reporting system to serve the many constituencies for process and production data.
Watch where the parts go
Gillette's CIM network starts with a Panorama system from Syscon-PlantStar of South Bend, Ind. It's what links all the molding machines (mostly Ferromatik models from 85 to 385 tons) and performs the standard functions of monitoring cycle times, injection pressure, injection speed, and melt temperatures. Whereas most molders' CIM systems would end right there, Gillette has integrated its Hekuma robotic systems for parts removal, orientation, and packaging into the network. With some help from its equipment and CIM vendors, Gillette created a link between Panorama and the Wonderware supervisory system that runs the automation.
Why is automation monitoring so important to Gillette? Its molding processes are already highly stable, says molding manager Wally Mallett, so automation accounts for up to 30% of the few unscheduled process interruptions that remain. The monitoring system helps Gillette both track the causes for that downtime and also fix any problems. Mallett explains that the PlantStar system counts all sorts of robot faults as well as the rejects that result. "It does a good job of providing a general tally of robot performance," he says. When a robot or other piece of automation goes down, a technician gets the specific diagnostic information needed to make repairs from a color-coded Wonderware screen at the molding press.
Monitoring the automation systems has given Gillette a better handle on scrap, too. Mallett recalls times when improper automation sequencing or timing had prevented the robot from keeping up with the molding machine. Any parts that the robot couldn't handle were diverted into the reject chute. On some lines, the automation-related scrap costs had reached $5000 a day, according to Mallett. "When you mold as many parts as we do, scrap can eat you up in a hurry," he says. Data from the automation system gave Gillette engineers the clues they needed to adjust times and sequences. "Sometimes it's just a matter of adding a few pauses between automation steps," Mallett explains.
Gillette uses the Panorama system for parts counting, but the job is performed more accurately with the aid of a link to the automation system. For example, a single touch-screen button on the Wonderware terminal can turn off individual mold cavities - not an uncommon occurrence when most of Gillette's tools have at least 16 cavities. Pushing that button tells Panorama to adjust its part count per machine cycle. Likewise, when parts must be diverted due to an automation problem, the link between Panorama and Wonderware fosters an accurate production count and lets Gillette tally the scrap.
Gillette also created a link between Panorama and its automatic guided vehicle (AGV) parts transporters. Since Panorama counts production, it made sense to have Panorama call an AGV in time to carry parts to the warehouse. The pains Gillette took to ensure an accurate production count pay off here. "If there's any missed production, the AGV won't be there before it's needed," says Mallett.
The Panorama system also passes the accurate production counts over to Gillette's enterprise resource planning system (MFG/Pro from QAD Systems in Carpinteria, Calif.). The ERP system, in turn, updates the radio-frequency tags that reflect the true count in each pallet of parts.
Making CIM simple
Gillette's large molding operation and massive production volumes could easily pile up an overwhelming amount of monitoring data. "It's easy to get bogged down in too much detail with 105 presses in your shop," says Estrada. He notes that Gillette employs two strategies for squeezing maximum benefit from its CIM data.
Challenged by differing requirements for production data across its organization, Gillette established an Oracle database from which a variety of custom reports can be created. "People want to look a their own piece of the pie," says Estrada.
Another Gillette strategy is to summarize the shop-floor data for each of the 45 product families at the Boston operation. For its daily "how-are-we-doing" meetings, managers focus on the Production Performance Index (PPI), a benchmark calculated by the PlantStar system. "It's really the theoretical maximum obtainable for a given product family," says Mallett. PPI encompasses a variety of factors, including cavity utilization, scrap rates, and cycle time. "No matter how complex your manufacturing system is, PPI works as a way to summarize your performance," Estrada adds.
Gillette shoots for at least an 80% PPI. "When we hit that, we're basically happy," Estrada says. That goal is tough to meet, considering that it requires at least 96% utilization of cavitation, 92% uptime, and less than 3% scrap.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Plastics Technology's 13th Annual CIM Leaders Awards: Take a Wide-Angle Look at Your Process|
|Date:||Dec 1, 1998|
|Previous Article:||Where's your next plant?|
|Next Article:||At Polymer Conversions Inc.: a controlled environment is critical to quality.|