Related: Automotive Powertrain
Historically, agility is thought to be nice to talk about as something that would be a characteristic of a company's production operations... at some point in the future ("We will be an agile company!" goes the proclamation). Or as something that academics and consultants use as a means by which they can cash in on the buzz through conferences devoted to the subject or through the development of a practice dedicated to the notion (and thereby create the opportunity for even more people to proclaim, "We will be an agile company!").
There are the random examples of companies that have actually implemented agile production systems, but these companies, in addition to their rarity, tend to be those that are making either comparatively small products or not a whole lot of products, all of which is to say that the phenomenon is more talked about than realized.
When it comes to volume production of comparatively large volumetric parts—and let's cut to the chase and admit that the subject here is automotive powertrain—the way of making engines has been—and in all candor, continues to be—through the use of dedicated transfer lines. The efficiency of this equipment has, over the production of millions of blocks and heads, been worked long and hard, almost to the degree that one could argue (and vendors of this type of equipment often do) that transfer lines just make the greatest amount of sense from an economic standpoint.
While the investment numbers as typically calculated still give the nod to dedication over agility, there are changes afoot. This is being driven in no small part by the end of a complacent customer base, which is giving way to one that is getting more fickle, and by requirements (legislative as well as market-driven performance-based) that are necessitating changes on a comparatively frequent basis, something that dedicated equipment doesn't lend itself to making. Although change may be no more appreciated today than it ever has been, it is also comparatively non-negotiable.
|Although this V6 is presently used in Oldsmobiles—and given what is going to happen to the Oldsmobile brand—the equipment that is being used to machine the components is agile, which fundamentally means that even if GM Powertrain stops making the engine (although it is more likely that they simply use it in models other than the Intrigue and Aurora), the CNC machines can be used in other applications.|
So some companies are working toward becoming more agile in powertrain production. General Motors is certainly one of them. And Robert M. Agresta, process manager, Manufacturing Engineering, GM Powertrain (Warren MI), explains what they are doing. He's spent 27 years in manufacturing. Given his age, he can probably go for about 20 more. In the context of that, this statement is rather telling: "My guess is there will come a day when transfer lines will be a rarity and CNC will be the baseline. It will happen before I retire." That point in the future isn't all that far away.
All Along the Learning Curve.
Agresta admits that CNC machining is nothing new. It's been around a long time. But it was happening mainly in job shops, with the occasional production environment, too. "GM started off with CNC producing components," he says. Things like water pumps. Front covers. But then, he notes, "It found its way into engine and transmission manufacturing, more for doing subsets of operations: the crank lines were probably the most intensive, with CNC milling, turn/turn broaching, CNC grinding and chasing pin-type activities. But it really hadn't made its way very much into prismatic parts machining, blocks and heads—the meat and potatoes of the metal removal function—until about 5-6 years ago." As with doing practically anything new and different, things didn't run smoothly from the start. And Agresta candidly admits it: "We endeavored to put in a complete unit of capacity that would do CNC manufacturing for cylinder heads. It was not as successful as we'd hoped. We endeavored and tried again. We put in some block manufacturing capacity. While it was better, it still had some opportunities. It was not quite as fast to market as we hoped it would be. We took the lessons from that and applied them to the V6 capacity." And the premium V6 engine that he's referring to, which is being manufactured in the GM Powertrain Livonia plant, is the unit available in the Oldsmobile Intrigue and the Oldsmobile Aurora.
Here is a Key Benefit of Efficiency.
Although it is clear today that the aforementioned applications of the premium V6s are not exactly, well, ideal, given the fate of the division in question, when Agresta and I talk, the announcement of the slow demise of Olds has not been announced. However, there is a distinction about agile equipment that makes such losses less difficult to recover from (at least from a capital equipment point of view): it is redeployable.
Agresta says there are basically four approaches:
A differentiator between these is how long it changes to covert from one type of product being produced to another (and he's basically talking engine-to-engine, not engine-to-toaster or something similarly distant, as sometimes cited by the academics and consultants when they talk about agility). In his estimation, it goes like this for changes:
- Agile—3 to 6 months
- Flexible—6 to 12 months
- Convertible—>12 months
- Dedicated—12 to 18months
Obviously, there is something to be said for the speed of changing over from one product to another—something valuable to be said for it.
In addition to which, he points out, "Agile means volume and capacity flexibility because as the system needs to grow or shrink, I can redeploy the assets from that system to other things." For example, they are building the premium V8—a.k.a., the Northstar—in Livonia, as well (in the case of the Northstar, it isn't an agile system doing the job but a transfer line with the addition of a bank of CNC machines added in the middle to provide flexibility for mounting faces or the movement of key surfaces. These machines, it should be noted, had been purchased for another program at another plant that dried up: "We saved that capital, whereas if it had been a transfer machine built and dedicated, it probably would have cost us 70-90% to retool."). If the demand for the V8s were to grow significantly, that V6 equipment could be used to do that job. Or maybe if the V6 goes away with Oldsmobile, the machines will just go somewhere else.
Agresta comments, "I don't have to expect that an asset spends its entire lifetime on a single product." CNC machining centers mean that as demands change, the means by which new things can be produced can be accommodated without having to incur major cost penalties.
Before the End.
Of course, that is an advantage after the fact. But having CNC-based agility provides plenty of upfront advantages, things that aren't necessarily obvious. For example, Agresta points out, "Preplanning accuracy is less critical." Which is certainly important given the drive to shorten design cycle time, because that means that Manufacturing is given less time to assure that the process to make the product is validated and responsive. "We can add and subtract features that come up late in the program through the design cycle. We can change from one diameter hole to another often with no additional tooling cost because we're making another hole somewhere else of that size and it is simply a matter of reprogramming what tool you select," he says.
|GM Powertrain has found that the use of machining centers in its plants is allowing smaller economic lot sizes to be produced. The machines can be sourced more rapidly than transfer lines, and they can be up and running faster, too. So instead of buying projected full capacity, they can buy the equipment that is needed incrementally. (Photo courtesy of Makino Inc.)|
Then there is the issue of determining what resources will be needed to do a particular job. Although product planners have a multitude of tools that they can use to predict the required volume of units to be built in order to meet anticipated demand, fundamentally, a prediction is a prediction: there are all sorts of variables that can make what would seem like a winner into a way-behind also-ran. But that can be determined only after the fact—after the time, effort, energy, and money have all been invested.
Agresta explains that machining center-based cells allow the incremental ramp up of production capacity, whereas transfer lines require, essentially, that 100% volume is purchased at day one—and it may be that it is never required (although it has been paid for).
Do It Right.
One of the things that Bob Agresta says is vital to agile production is a fully capable tool room, where tools are preset and I.D. and offset information are loaded into a memory chip that accompanies the tool before, during and after its use. Agresta explains, "We want to cut it"—as in "the part"—"right the first time." There's no time for the cut-and-adjust approach that has long been common to production operations. Operational efficiency of the production equipment is essential, so it makes sense to have everything prepared ahead of time so that the equipment can do what it is meant to do, which is to make good parts.
The fact that he cites the tool room as being important underscores the fact that agility requires more of a holistic approach to the operation, recognizing that there are effects of various other elements of the system on getting the job done. So prepared tools just make good sense.
Learn One; Know 50 (or more).
When it comes to maintenance, there is the benefit of consistency and similarity: there is plenty of the same equipment on the factory floor.
Agresta pauses at a section of a transfer line for the V8 production at Livonia and points out that in this one particular area there are three completely different motors. If there is a problem with one of them, then it is probably the case that the maintenance person will have to resort to manuals (which is probably going to be challenging with regard to finding the right one) to discover the intricacies of that particular motor. (And then there is the issue that if a station in a serial transfer line goes down, then due to the sequential nature of the line, the entire line is affected, which means that the motor must be gotten to ASAP—if not sooner.)
As Agresta points out, at Livonia there are 55 identical machines. What the maintenance personnel learn about one of them is applicable to the other 54. Consequently, there is a strong likelihood that downtime diagnosis and repair can be greatly expedited as compared with a transfer line ("Hmm... What type motor does this station have... ?"). What's more, if one of the machining centers goes down, if the system is set up with a parallel architecture, where there isn't a direct dependence of one machine on another, then part production can continue (although it may be slightly reduced because of the fractional loss of the single machine).
The Most Important Asset.
One of the tendencies that people might have when considering a move from traditional mass manufacturing to an agile approach is to think about the equipment: the machine tool, the tooling, the fixturing, the material handling, the gaging... All of which are important, of course. But Agresta points out that this is an insufficient consideration. In fact, so far as they are concerned at GM Powertrain, when it comes to an agile system, fully half of the time that is spent in the development of the program should be people-oriented. The other half is for the equipment. "People issues shouldn't be ignored, but looked at as an opportunity. We were very concerned that the amount of training was going to be insurmountable. To that end, we took our folks to the machine tool builder to run machines to build prototype parts. Had them learning about failure mode affects, how to detect when a tool was wearing out, building control and operating plans rather than sitting around and guessing. Engaging the people in the process who are actually going to do the work: a tremendous benefit—they became mentors and teachers. Their ideas were as valid as anybody's: a lot of knowledge about transfer systems and quality checks and measurement systems that we could have only guessed at—they knew the techniques, the implications of being right."
And given this learning and the innate adaptability of people, not only has an environment been created that is agile in terms of production equipment, but also agile production people.