Cells: The building blocks of manufacturing:How to manufacture manufacturing cells(Web-only expanded content)

If the future of molding in the United States is in systems and cells, then learning how to put them together is a critical part of survival. Given all of the processes and requirements involved in cell construction, what’s the best practice?

Automate, or else, says one molder. “Most molders are slow to make the commitment to full automation,” he says. “Most are trying to get by with standard purchases of stock molding machines and robots, or parts pickers, or conveyors, and maybe a boxmaker to reduce labor content.”

This source says these molders are still thinking about labor reduction instead of labor elimination.

“Mold the part, remove it from the mold robotically, perform any secondary finishing or assembly automatically, inspect it with artificial vision, and put it in the box. No labor—full automation,” he says.

“You can’t do all this without a dedicated system built specifically for the product you are going to produce, and you can’t use this setup as a multipurpose system. It has to be dedicated to one part, or possibly a family of parts.”

So where do you start? How do you assemble a cell? If it all starts with part design, it starts with the mold, right?

First Contact

One molder says it should start even earlier. “We have to involve the designer,” says technology solutions product manager Hermann Plank of Tech Group Americas (Scottsdale, AZ). “Based on the designer’s expertise, we pick the right technology to mold the part, or to assemble it.”

“The most challenging portion of any turnkey system should be done by experts in that area that also have diverse enough backgrounds to integrate an entire system,” says Donna Bibber of Miniature Tool & Die (Charlton, MA), a turnkey systems provider specializing in micromoldmaking.

“If the mold is the specialized area, the moldmaker with integration experience should be the cell builder. If the system challenge is in assembly or automation of several components, an automation house should be the initial focus for the cell design, and so forth.”

Integrating Systems

John Gravelle, president of Mar-Lee Cos. (Leominster, MA), is an accomplished medical, multishot, and IML systems integrator. Mar-Lee is also a moldmaker and a molder.

“Most molders do not have the capabilities to lay out, specify, and manage a complete system build-out. And if it’s not done in a very coordinated, overlapping manner, then there will be disappointments and finger-pointing between the vendors,” Gravelle says.

Mar-Lee itself accepts complete responsibility for system build-outs.

“If we are to be the molder, we require a long-term customer commitment, guaranteeing annual volumes with ‘if-not-achieved’ penalties.”

Ron Embree, president of Moll Industries Inc. (Dallas, TX), says, “At Moll we often do this ourselves and then manage the tool build. However, as our industry advances and becomes more specialized, so do the skills of moldmakers.” Moll reportedly has identified several moldmakers in North America and beyond who have specific skills that it chooses for appropriate applications.

Space Between the Bars

Raymond J. Veno, VP of engineering and technology at Precise Technology Inc. (North Versailles, PA), says moldmakers should be contacted first, especially if you want to spec the right molding machine for a cell.

“We must first determine the mold cavitation and size to meet the demand. That said, the cavitation layout and mold design are critical to overall system performance for part takeout, as well as for IML applications.”

Minimum in-and-out robot travel, as well as inmold secondary motions, must also be considered. “Mold opening strokes and EOAT profiles must be considered when specifying molding machine daylight and tiebar length factors,” Veno says.

The Holistic Approach

A key point all stress is that no single party should act as an independent vendor. “Don’t let the moldmaker get two weeks or more into the project and then ask him to change something like center distance,” Gravelle says. “He’ll want to get paid for the change.”

Dennis Tully of Miniature Tool & Die says that the better individual experts in the process, molds, and machinery work together, the quicker the system can be up and running.David B. Williams, VP of operations at Cascade Engineering Inc. (Grand Rapids, MI), agrees with the holistic approach. Once specific details of the product and customer requirements are known, Williams says a cross-functional team should be formed.

“Out of this process we are able to understand the critical requirements for manufacturability. It is difficult to say we always start with one contact first, as it often varies. Sometimes three or four critical variables are interdependent.”

Finger Pointing

Who should test, debug, install, and maintain the cell? Legally speaking, this is a very complex question. Usually a system up-charge is involved to cover such responsibilities, but sources say the up-charge is hard to estimate, and often even harder to justify. So, who’s ultimately responsible? Opinions vary.

“The sole responsibility should rest with the cell’s project manager,” says Miniature Tool & Die’s Bibber.

Cascade Engineering’s Williams says the responsibility rests with the supplier, or the system contractor—most of the time, anyway.

“It’s a price/value judgment. If the value of the service exceeds the price element, we are comfortable with controlling it ourselves. We have the experience,” Williams says.

Molders must accept overall responsibility, according to Mar-Lee’s Gravelle. “An independent moldmaker has no interest in accepting that level of responsibility. The machine guys may accept responsibility for the mold and then contract it out, but they rarely want to get involved with heavy automation.”

The Preferred Partner Martix

Conversely, Tech Group’s Plank says machine manufacturers should bear most of the responsibility in startup and training. Machinery OEMs have the best infrastructure, including crane capacity and floor space, he says.

“It’s also important that the team members have built a relationship through previous projects,” Plank says. “I call it a ‘preferred partner matrix.’ Each team member has a pool of vendors to choose from and, depending on the application, they can pick the experts they need.”

“This is definitely a supply chain effort that must be agreed upon by the stakeholders early and validated early,” says Precise Technology’s Veno. “Parochial interests must be set aside to get the common ‘best path forward’ established. We apply prescribed program management methods to ensure everyone has their roles and responsibilities covered in a working document.”

Moll Industries’ Embree adds, “This is a joint responsibility, since everyone has had input in defining the requirements and everyone has a stake in the application’s success.”

The Lead Time Quagmire

The lead-time-to-production-start equation is the biggest problem molders tell us they encounter when specifying equipment and getting cells running. Most say automation is the controlling factor.

The mold and machine reportedly arrive in around 16 to 18 weeks, but the automation can take two to three weeks longer to arrive. When you factor in a few more weeks for installation and training, molders say you’re looking at 20 weeks or more to get the cell up and running.If you have to be in production and can’t waste time waiting for the automation to arrive, you may have to tap into your human resources to do secondaries, inspection, and packaging. One molder who did so tells us his direct labor cost was more than $4500/week and his scrap rate skyrocketed to 8%. He didn’t give up on automating, though.

“Six weeks later we installed the automation, eliminating the labor. Scrap came down to under .5% and our cycle was reduced by 1 second, or 7%. It was easy to calculate the benefits and payback of automating,” this source says.

KISS

Other points to bear in mind for successful equipment integration, cell specifications, and startups include the following:

“Our experience has been to apply the KISS [keep it simple, stupid] theory when it comes to automation,” he says. “It’s better to add more stations than to try to do too much at one station and get overly complicated. Unfortunately, we learned this the hard way on some of our early systems.”