Printer Friendly

Machining cell integration: the role of the consultant.

Machining cell integration: The role of the consultant

The flexible machining cell is a great thing to have. Benefits include round-the-clock consistency, new levels of responsiveness, and reduced costs, inventory, and lead times. But getting there is not easy. Your first flexible machining cell, your first big step toward system integration, is often too big a leap to take on your own. If you can get all the help you need from the machine vendor, fine, but you may find that you need the objectivity of a third-party expert to put it all together and make it work the way you expected it to when you initially started out on this major endeavor.

To examine the role of the consultant in cell integration, we went to a leading integrator, Andersen Consulting, Chicago. This $1.3-billion arm of Arthur Andersen & Co has more than 13,000 professionals providing integration services, largescale computer design and installation, software development, and "change-management" consulting around the world. In October 1987, they opened a Systems Integration Center at Northwestern University's Research Park.

Mark Kuchel, director of that center, is a hands-on guy, presently in the field, working on a major multi-year client project and supervising a healthy number of people on various on-site projects, including several FMSs. His experiences from initial hand-holding to final certification provide valuable insight into the obstacles between the initial FMS idea and the final achievement of a working, productive cell.

What's a cell?

There are as many definitions of what constitutes a cell as there are people installing individual islands of automation. In Kuchel's experience, a cell is whatever you choose to bite off at any one time and fully integrate. "There is no common definition," he explains. "In some cases, it's just one or two related operations creating a part that's either stocked or moved to an assembly area. In other cases, it's a combination of as many as eight or ten operations, all tightly linked under the supervision of a single cell controller.

"So, whatever they want to call a cell is fine with us. It's a logical grouping of activities under one controller. Depending on the capability of that controller, this can include quite a few very different operations, or merely one or two. Not too many people have mastered the concept of area control yet, where you link multiple cells together under an area controller."

So, to avoid that challenge of a higher level of CIM complexity, people are simply redefining cells as something larger than the most elementary cell concept, yet smaller than a clear case of area control? "Yes, from what I've seen, because people are struggling with full coordination between cells, they have conveniently redefined cells as just bigger entities."

What's the big hurdle here, software? "Primarily."

So why not simply quit at the multiple-cell stage, if they're working well - where's the incentive to further integrate? "The integration of cells means streamlining them, coordinating them, eliminating any inventory between cells, and making the flow through (or flow time) between cells much more predictable."

Isn't this cell-linking step more difficult to economically justify than creating cells in the first place? "Yes, this is where a lot of people are struggling - putting it all together. If you have enough money in the company checkbook, buying an FMS is relatively easy. Linking it into your other systems is the real problem - tying the FMS-created parts into the assembly process downstream, for example; and making sure the raw materials for the FMS are delivered on time. Unfortunately, what often happens is a lot of inventory sits on both the in-bound and out-bound sides of the FMS, indicating people bought automation and haven't been able to integrate it yet. And with the cost of an FMS today, people are realizing they need to get more out of their investment - the need to fully integrate the cell into the factory process."

What portion of companies, in your experience, now have or are in the process of acquiring cells? "Assembly lines remain typically conveyor transfers with manual methods and little automation, but in machining operations, depending on cell definition and level of automation, probably 30 to 40 percent of manufacturing has some examples of machining cells with some degree of automation and integrated controls."

Parts made to order

What are the key application/selection criteria? "Low volume, of course, is one - something that requires only a couple days per month of production time: a parts-made-to-order environment, with a lot of custom work. Ideal FMS candidates, for example, are an industrial-equipment supplier making 200 units each day, but who needs only 10 units per month of one particular design. Or an engine manufacturer who needs to produce a wide variety of replacement flywheels for units in the field. Instead of doing batches of these parts and stocking them, you use an FMS to produce them on demand.

"High volume is still better handled by transfer lines or dedicated automation, because it will always be cheaper than paying for the flexibility of an FMS - particularly flexible tooling, the capability to make different parts on the same machine. The driving criteria is the flexibility that eliminates setups. Once the system is debugged (i.e., first-part right), you gain the uniformity of being able to make that part right every time thereafter. The program's there, the tooling's there, so the part can be produced on demand.

"The four- or five-axis machine required for an FMS can do a lot of different cuts, and the range of parts within the cell's part family is relatively wide. But you must pay for this flexibility up front, and some tradeoffs will have to be made.

"Another benefit is that you can also count on a good production history - the ability to track a lot of information on how and when each part was made. Also, introducing part-design changes is much easier and quicker - this can be a simple software modification with no hardware modifications required."

FMS modularity?

Must you always design the cell from the ground up, or will we ever reach a point where creating a cell will be simply a matter of selecting modules, with minor custom modifications of tooling and software? "Not for a while, I'm afraid. There aren't really that many FMSs out there, so while we're seeing some custom machines use modules developed for use elsewhere, most software is still custom designed to meet unique customer requirements.

"Sure, the simple cell may become a commodity, but most people have special situations. More typical is the five-axis machine, 100 or more tools, and a part family of 100 different parts - i.e., a complex environment. And this is where people are investing right now, not in the simple environments.

"We've been involved heavily in linking an FMS into higher level systems, either at the area or plant level. Typically, these cells are self-contained and do not communicate with the outside world. There is a lot of information that needs to be downloaded to the cell - the production schedule, for example - and the production history needs to be uploaded. There's a role here for someone, and typically, one the machine-tool builder or FMS vendor is not that willing to play. They're being asked to be cell-control software vendors, and that's not what they really want to do."

Software reruns?

How much of today's software is custom? "Vendors, of course, save software-development costs by rolling their software into the next similar application whenever they can, but our experience is that not much software we see has been used before, or if it has, is not typically a true package and easily maintainable.

"There is always the question of who's going to maintain cell-control software. A number of my clients have struggled with that problem. In some cases, the only people who can maintain it are vendors, and in a lot of cases, they don't really want to. For the user to be able to go it alone implies good modular code, software that's easily maintainable, well documented, accessible, etc. I've seen situations with only object code because the vendor doesn't want to leave them with source code."

When to call for help?

At what stage are people turning to consultants for help? "In some cases, we get involved in the early evaluations, system design, and selection of an appropriate FMS vendor. More typically, we get involved after the FMS is purchased and present, and they are trying to debug it and link it to the rest of the manufacturing process. This is when they recognize it's more complex than they thought, and realize they need the help of an integrator.

"To get the real benefits from any FMS, they must tie it into related processes, or integrate existing cells into a higher level or area-control approach. When they realize they are dealing with a lot of new technology, when their MIS guys realize its a new league for them, and when their engineers realize how much software development is involved, they conclude that there's a void here that needs to be filled by an integrator.

"More and more, the role of the integrator is as a company with a large set of skills that can fill gaps wherever necessary. Depending on the skills that exist in the vendors you're working with, how willing they are to get dirty in solving the problems, their on-site installation-support capability, plus the client's capability - these factors will determine what voids need to be filled. It's very important for the integrator to have the ability to fill those gaps as necessary."

There's also the benefit of going to a third party when you feel you're in over your head and concerned about being led astray by a vendor trying to push his own product - whether machine tool or software - instead of what's really best for you. "Yes," he laughs, "objectivity seems to be selling quite well lately. Of course, everyone wants to expand their role and capabilities - the typical software houses trying to move down the pyramid and machine-tool vendors trying to move up. As more companies realize the important role of an integrator - that they have to have one, whether they want one or not - we're in an ideal situation with a very healthy future."

Cost for integration

How much of system cost is the cost of the integrator? "The key factor in cost is how much flexibility is required. Loosely defined, a cell can range from as little as $50,000 (one machine, and our only involvement was in determining what kind of control was necessary) to $15 million (multiple machines, AGV part delivery, downloaded tooling and NC programs, AGV transfer between machines, automatic inspection, and lights-out operation). And our role in this varies just as widely, but generally the total cost of integrating hardware and software (machine vendor software plus our own contribution) will be in the range of 10 to 20 percent of system cost. The machine-vendor capabilities will, of course, dictate the role of the integrator and the proportions for each."

Tooling implementation

Kuchel says he gets nervous when people ask about modular tooling and modular fixtures. "All those come with a price tag for this added complexity. That seems to be the `in' thing right now: fixtures you can use anywhere and tools that are flexible. But this pushes you toward a more complex environment that demands more flexibility, complicates the software, creates additional debugging problems, etc. And this complexity makes manufacturing people nervous. They're used to machines they can work with and fix, where in the typically sophisticated FMS environment, a large amount of training is required and higher skill levels are involved."

Upgrading the workforce

What's your role in upgrading the workforce? "We are involved with a little of the training. That change process is a delicate one - especially in a heavy union shop. This is not usually our problem, but it's part of the overall communication process - making people understand what this new environment will entail and what to expect. We do get involved in helping define that new world for them. The more communication up front, the fewer problems down the road."

Are people really planning ahead - visualizing these problems before they occur? "Most people are well aware of what's required to justify an FMS, but they're not always willing to do what's required to make it work right. These are million-dollar systems. People usually do some homework before investing that much - they are not typically naive, but they haven't really thought it all through either. Our job is to point out these things ahead of time so that they will have more options on how they ultimately choose to attack these problems."

Measuring the advantages

Based on your experience, what are the key, measurable FMS advantages people have realized? "Increased throughput, reduced setup, and improved quality, and these can be quantified. Of course, the initial justification process comes down to either you believe in FMS or you don't. We've seen ROIs well above 20 and 30 percent per year, and that should be your target. Otherwise, why invest? If you can't beat 15 percent, find another place for your money!

"So, until the justification process gets more refined, you must decide whether to take the optimist or pessimist view. So, basically, that one third of manufacturing who are optimists have taken the leap into FMS, while the remaining two thirds of us in this country who are pessimists have not, and continue to fall behind this technology."

Concludes Kuchel, "I haven't had much success in convincing people who don't want to go forward that they should." The technology pessimists will continue to keep their heads in the sand.

PHOTO : A tool robot loads the back magazine of a Kearney & Trecker 600 machining center from a palletized tool kit. The cell controller checks the tools in by laser identification and already knows the sequence in which they will be used. It also knows in advance what machining-cycle modifications to make for each tool's minute deviations from specifications.

PHOTO : As cell users become more experienced and know better what they need to sustain these systems, they will demand more responsive and flexible software.

PHOTO : A great training ground for new people, Andersen's System Integration Center is also working on flexible assembly systems for electronics applications.
COPYRIGHT 1989 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Sprow, Eugene E.
Publication:Tooling & Production
Date:Nov 1, 1989
Previous Article:Needed: more technological commercialization.
Next Article:Stampers turn to transfer presses.

Related Articles
Show report: 6th EMO - panorama of FMS.
Flexible fabrication - a progress report.
Smart management seeks flexible manufacturing.
Off-the-shelf cell control.
Flexible machining systems & cells.
Automation swings to cells.
Machine cells meet JIT need.
Flexible manufacturing expands its envelope.
PCs link offices to saws.
Flexible manfacturing for auto parts. (Machining Centers).

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters