PART 4 of 6

Follow this series oF "World Class Vehicle Launch Timing" as it appears in the magazine... Part 1 Part 2 Part 3 Part 5 Part 6

We identified a total of 10 vehicle launches by Chrysler during the period investigated. Six of those launches met the single product, single plant requirement for this investigation. Our discussion focuses on the performance of these six launches.

Figure 1: Product Launch: Percent of Capacity-Chrysler, World Class and Industry Average, 12 Months Prior and 12 Months Following Vehicle Launch.

The data for this article was gathered from Ward's Automotive Reports. For each launch, a production start date was determined. Monthly production data was collected for the 12-month period prior to the start date, and the 12 months subsequent to the start. We compared this data to Harbour & Associates (Troy, MI) capacity estimates to facilitate direct comparison between plants of differing capacities. Annual capacities for each plant were gathered from The Harbour Report, and similar tables published in Automotive News. These estimates were divided by 12 to give a monthly capacity estimate for each plant. They were then divided by actual production to give a monthly performance to capacity measure. Figure 1 compares Chrysler's performance to the North American industry average and to World Class, the company with the quickest vehicle launch. The chart shows that, on average, Chrysler facilities demonstrate a decline in production at changeover more severe than the industry average, and they take somewhat more time to return to a normal operating range. Chrysler facilities return, on average, to a higher capacity utilization rate post launch than the industry average.

Table 1 shows the six Chrysler plants and launch events that met the single plant, single product criteria. (Author's note: Warren Truck was treated as a single plant, single product plant even though the Dodge Dakota is also produced at the facility.) Of all North American participants, Chrysler may have had the most significant amount of change—both in terms of products and location. From the amount of change in its light vehicle line-up, it is apparent that the organization was not the only thing completely re-engineered in the early 90s. This change is indicative of both the distance between Chrysler and its competitors at the start of the decade, and a newly established goal to be the industry's style leader.

Table 1: Selected Chrysler Plants and Launch Events
Plant	Launch Date	Product
Bramalea	January 1992	LH sedans
Warren Truck	May 1993	T-300 Ram
Belvidere	June 1993	Neon
Sterling Heights	April 1994	JA sedans
Windsor	June 1995	NS Minivan
Pilette Road (Windsor)	July 1995	Ram Van

Figure 2 graphically shows the launches reviewed for this article. All facilities produced no vehicles during the initial month of the launch. Belvidere, Bramlea and Sterling Heights experienced several months of downtime between the end of production for the previous vehicle and the beginning of significant production of the new vehicle. Each of these facilities required significant change in equipment and tooling during the changeover. Since the authors choose to measure the new vehicles launch from the time the facility stopped producing the old vehicle, the extended downtime of these plants made the launch curve somewhat flatter than if we measured it from the month of when the first vehicle was produced. However, given the performance of world class launch companies, we felt that the clock would start running the moment the previous model production ends.

Last fall, Chrysler launched two key products, the replacement to the LH and the Dodge Durango. These launches allow for a comparison to earlier launches—at least in terms of the one measure examined in this series—that of `ramp-up' time. The company's recent launches appear to follow the same pattern as those of recent years, taking six months to return to capacity. The launch of the Intrepid and Concorde was reportedly slowed due to a supplier's inability to meet the launch schedule.

As has been pointed out many times in this series, we have investigated only one aspect of the vehicle launch process. Does Chrysler's apparent lack of progress in decreasing the time to return to full production indicate a differing philosophy?

Figure 2: Percent of Capacity-Selected Chrysler Facilities 12 Months Prior and 12 Months Following Vehicle Launch.

"The real measure of a launch is `Did you meet your goals?'" said Frank Ewasyshyn, vice president-Advanced Manufacturing Engineering at Chrysler. According to Ewasyshyn, the company sets goals for each launch, yet they realize that trade-offs are required. Ewasyshyn identified quality, quickness, and meeting the objectives all within the investment parameters as the true measure of any launch. "You can throw money at a launch, and get what appears to be a fast, great launch, but it probably won't be cost effective."

The platform teams are responsible for vehicle launch at Chrysler. However, Advanced Manufacturing Engineering supports each platform team. It serves as both a staff and core function within the organization. Representatives from advanced manufacturing are on each platform team, but advanced manufacturing also maintains an independent staff dedicated to maintaining critical skills. Knowledge is shared between platforms, not via tech clubs, but instead a more formal review board consisting of platform team members and advanced manufacturing representatives.

At Chrysler, a company that has fought long and hard to improve its quality reputation, the search for quality drives each vehicle launch. "Quality is the sole and only gate for a launch at Chrysler," Ewasyshyn said. "Others at Chrysler have said it, and I agree with them; everyone will remember a quality product, but few will remember a fast launch."

Ewasyshyn pointed out that a fast launch may, in fact, have some drawbacks. The level of inventory of the model being replaced plays a role in beginning the speed of a launch at Chrysler. Ewasyshyn likened the strategy to keeping a pail with a leak full of water, suggesting that you have to pour water in at the same rate it leaks out, or you will throw off the equilibrium. "You have to look upstream. If you have seventy-five days inventory of the old car, you may not want to get the new model out quickly. If you do, you'll need huge incentives to move the old model," said Ewasyshyn. "The ability to turn a plant quickly is desirable, but you don't always need to use it.

Another key strategy at Chrysler is to provide exciting products. This commitment to styling may present trade-offs at vehicle launch. Many suggest that in order to make more rapid changeovers, companies need to maintain engineering limit styling changes between the new vehicle and the one that it replaces. "We would never go to the designers and say hold the hardpoints so we can have a faster launch. Styling is a core value at Chrysler," said Ewasyshyn.

Figure 3: Chrysler 1998 Model Year Vehicle Launches.

Chrysler is working to shorten launch time within the aforemen-tioned parameters. The company has increasingly been using virtual manufacturing to analyze the feasibility of manufacturing processes. "Virtual manufacturing is part of getting to your stated objective," Ewasyshyn said. "We do simulation verify, but you need to be honest in analyzing the simulation."

In an attempt to move closer to a complete virtual manufacturing system, Chrysler recently announced the Control Program Generation and Analysis (CPGA; see pg. 54 for the complete story). CPGA builds on Digital Manufacturing Process System (DMAPS), a virtual manufacturing system that Chrysler introduced three years ago in cooperation with Dassault Systemes. DMAPS now provides Chrysler with a fully computerized end-to-end product and process management system which enables Chrysler to design, construct and run a "virtual manufacturing process." The project is an example of Chrysler's Extended Enterprise concept. Along with Dassault, Rockwell Automation, Deneb and Progressive Tool & Industries were partners in the development of the new system.

According to Ewasyshyn, CPGA will reduce the time it takes to program a typical workcell by thousands of hours, shave two months to four months off the development time of passenger vehicles and save upwards of $20 million per assembly variation in the build process, leading to better vehicle quality. CPGA is also expected to improve communication among Chrysler's manufacturing and engineering, and the members of the Extended Enterprise.

Chrysler has frequently stated that it wants to be the world's premier automobile company in the coming decade. Naturally, it is reasonable to wonder if being relatively slower than their competitors may preclude them from achieving that goal. In response to that, Ewasyshyn said, "Chrysler views the goal of being the premier automotive company much like a competitor in the decathlon; the winner is usually outstanding in all aspects of the business, but not necessarily the best in any one."