PART 6 of 6
The first five articles raised numerous questions and interesting insights regarding the strategic importance of a new vehicle. As we introduced this series, we raised the theoretical question of a marginal cost curve for vehicle launches. In the model, we suggested that each vehicle firm explicitly or implicitly define its manufacturing cost schedule for vehicle launch times. We also theorized that the companies must make strategic decisions with regard to capital allocations. These decisions would cause a company to move along its launch cost curve, but only through a change in technology or process could a company shift its cost curve. We also assumed that companies face fixed schedules for launch costs and that there are differences across companies (below). In retrospect, this model appears to have some validity. The companies investigated do appear to have very different cost curves—although saying this with complete certainty would require a Ken Starr-like investigation!
One element of our original theory seems even more applicable now than at the outset of the series. Our marginal cost model of launch performance assumed that the manufacturing cost of launching a new vehicle generally fell with the amount of time devoted to the launch. This is rather counter-intuitive when one considers the mounting cost of an idle work force and unused facility capacity. Upon further investigation, it appears to be true that it is generally more expensive to shorten equipment installation periods of time within a company's launch cost curve (or paradigm). During the past several years, several companies appear to have attempted to shift their cost curves, when in fact what they may have done is merely moved along their original line.
Another question we raised as we entered this project was the level of change the product undergoes. A review of the launches indicated several of the platforms were as much as a decade old. A facility producing an older platform will need significantly more investment and associated downtime than a facility producing a much newer platform. This is in part due to the tendency to neglect an older product and the tooling associated with the product.
The launch of a new vehicle is in many ways the ultimate test of a good coordinator. Not only must the assembler coordinate the development of the vehicle and the conversion of the assembly plant, but they must also concurrently work with their suppliers to ensure quality and volume at Job 1. Jerry J. Harvey, Executive Director of the Manufacturing Engineering, Operations and Integration Center (MEOI) at General Motors pointed out that, "It is important to remember that a vehicle launch is not only an acceleration at the assembly plant. Most suppliers are also undergoing an acceleration process at the same time. It is important for engineering to interface with the suppliers to assure that their new process technology is in place."
Chrysler, the company with the poorest performance with regard to launch speed, may be close to realizing great gains in the near future with regard to supplier coordination and the launch process. It has been well publicized that several of Chrysler's recent launches have had delays due at least in part to supplier difficulties. "Chrysler's suppliers are coming up to speed on CATIA, and as the suppliers become more adept at managing the software, they will become a huge asset for Chrysler," said Mike Tracy, president of The Agile Group, Inc., a consultant with extensive experience in the program management of product—both automotive and non-automotive—launches. "Chrysler's commitment to share design data with suppliers upfront has made a huge impact," he added.
"Our recent project with the Packaging Systems Division of DCT for the 1998 LH was an indication that Chrysler is willing to share design data early in a program," said Tracy. "DCT needed the data to effectively design the racks, dunnage and material handling systems at Bramelea. That up-front sharing will really be visible in coming launches."
The coordination of a launch can be made even more difficult if a supplier is concurrently launching a new facility. This past summer, Chrysler began production of an all-new Grand Cherokee, with an all-new body shop, a new engine from a new engine plant and a new transmission from a new transmission plant. "We wanted to make a real statement," said Frank Ewasyshyn, Vice President-Advanced Manufacturing at Chrysler. "It is a reflection on the importance of the Grand Cherokee. We intend to position this product as a technology leader, so we had to introduce the new engine and transmission."
The launch of an engine or transmission plant presents major timing challenges. Modern engine and transmission facilities have large amounts of highly automated machining equipment. This equipment is necessary to meet the high volumes and high precision required, yet it requires a higher level of fine tuning. Beyond the mere effort to tune, or warm-up, the head and block machining lines, and the engine and transmission assembly lines, there is the so-called high infant mortality rate challenge. That suggests that with machine tools there is a likelihood that breakdowns in highly automated machinery will quickly surface in the first few months of operation: after the `bugs' have been worked out, the machinery has a tendency to operate relatively problem-free for extended periods.
However, in this case, the synchronous launch of an assembly plant may in fact give Chrysler a launch cushion—at least for the engine and transmission plant. The New Kokomo transmission plant first started trying out their production equipment at the plant in March 1997 and began volume transmission launch in April 1998, several months before Job 1 for the new Grand Cherokee. Beyond a slow, cautious launch curve for the engine and transmission plant, Chrysler has the luxury of a proven back-up. A majority of the 1999 Grand Cherokees will have the carry-over six cylinder engine and transmission, thus allowing the company to cover any snags that may occur in the production of the new powertrains.
Yet, even given the immense challenge faced by the assembler from the coordination standpoint, the flexibility to quickly change tooling in a body shop remains possibly the most critical—and if done poorly—the most costly element of a changeover. In this measure, Honda appears to be fundamentally ahead of everyone else. According to Tracy, "Honda does it more efficiently than the others. Their vertical integration with R&D North America (HRA) gives them an advantage. They also have shown a discipline to keep consistent vehicle architecture."
As pointed out in previous articles, the essence of the difference is that Honda appears to be the only manufacturer that does not use some form of "flex space" for a new vehicle launch. Flex space can be described as the practice of assembling new tooling beside the current assembly line to allow time to tune up the new equipment prior to launch. Conversely, Honda started building pre-production 1998 Accords on-line in March 1997. The flexibility to quickly move tooling into (and out of) position allowed Honda to gain first-hand experience, without interfering with 1997 production. That kind of flexibility allowed for a launch that happened over a weekend. It also may have led other manufacturers to increase their use of flex space to quicken their launch curve—albeit at a likely significant cost penalty.
As this series developed, it became apparent that the companies had different priorities regarding the launch of a product. Yet one point remains clear—the cost of a slow launch can be huge. Tracy put that cost into perspective. "The contribution margin on a $20,000 vehicle, once variable cost and supplier cost are accounted for, may be in the range of $11,000. No matter what a company says its launch priorities are, the only measure that counts is the total corporate bottom line. Some companies spend more overall in launch because they spend less elsewhere (e.g. in the development of the vehicle) as a part of overall corporate savings equation. Also, the fact that companies are willing to slow the launch process down to ensure quality is an indication of the competitiveness of the market."
The 1999 model year presents several interesting case studies. When we highlighted the GMT-800 launch last spring, we had no idea that the program would be held hostage by a nearly two-month long strike. Yet, according to General Motors, the launch is proceeding at or ahead of schedule. The Grand Cherokee launch will surely present some interesting lessons, and Ford will soon launch their long awaited DEW 98 program, with an all new product, engine and transmission. And of course, Honda is launching an all-new minivan at its new Alliston, Ontario, facility.
Finally, we wish to thank those companies that were willing to give some insight into their launch strategies. When this series began, we were warned that this topic, maybe more than any other, may place individual companies in a less than positive light. After a year of investigation, it is obvious that the challenge of launching a new vehicle is something that each company is keenly aware of, and not necessarily willing to discuss. We greatly appreciate all whom assisted in this project, especially Jerry J. Harvey of General Motors, Frank Ewasyshyn of Chrysler and Mike Tracy, of the Agile Group. Their insights and time added greatly to this project.