Accelerating Product Development for Diesel Engines

Article From: 10/1/1998 Automotive Design & Production, , Editor-in-Chief from Gardner Business Media, Inc.

Here's how Detroit Diesel Corp. went from a clean sheet of paper to a fully functioning diesel engine in just 7-1/2 months. Their experiences can provide you with some solid lessons.

This could be a world's record.

A cross-functional team at Detroit Diesel Corp. (DDC; Detroit)—one consisting of engineers of all relevant varieties from both DDC and an array of supplier companies—went from the proverbial and actual clean sheet to a fully functioning diesel engine in 7.5 months. 

Diesel engine development
This graphic gives some of the fundamentals of rapid product development performed at Detroit Diesel for the DELTA engine: (1) Do a quality function deployment so that market desires or requirements can be translated into technical characteristics. Once identified, then make sure that they are adhered to through the program. (2) Implement advanced technology, including such things as computer-aided design and rapid prototyping. If your suppliers have these resources, use them. (3) Make sure that manufacturing people are involved in the team so that there are as few surprises as possible when the real thing needs to be put together.

That's right. A new engine. Specifically, a 4.0-liter, 60o, V6 engine that's targeted at the North American light truck and sport utility vehicle (SUV) market. DDC, which does have a line of automotive diesels ranging in size from 125 to 160 hp, power plants that are utilized by automakers primarily in Europe, needed a clean sheet approach because, according to Charles E. Freese, V, director, Automotive Sales, and the man who, in his previous position of chief engineer, Advanced Automotive Diesel Engines, led the team, the North American duty cycle for light truck and SUV engines are significantly different than those in other parts of the world.

The American Road


Charles E. Freese V
Charles E. Freese V headed up the development of the 4.0-liter V6 engine program. Prior to that, he headed up the team that designed and developed the diesel engine that is used in Chrysler's ESX2 hybrid electric vehicle.

"We couldn't use a derivative engine," Freese remarks, "because the North American market has unique requirements." And as an example, he cited personal experience, heading north on I-75 in Michigan on a summer Friday afternoon in an SUV towing a boat. As anyone who has made that trip knows, the travel is stop-and-go, bumper-to-bumper. Freese notes that although diesels account for about 40% of the auto engines in Europe, those engines don't deal with conditions like that. The duty cycles are, indeed, different in North America.

What's more, there was the added concern that in North America, automotive diesels don't have much going for them from the standpoint of reputation (e.g., hard to start, noisy, smelly).

Management at DDC and parent company Penske Corp. recognize that the number of light trucks (i.e., the ubitiquitious pick-ups) and SUVs being sold in the U.S. is such that the OEMs may be facing an economically demanding situation if the Corporate Average Fuel Efficiency (CAFE) levels go up: these vehicles may represent high profitability for their builders, but they also represent low miles per gallon. So the DDC folks reckon that given the 30 to 50% higher efficiency of a diesel engine compared to a gasoline-powered one, the North American OEMs might be inclined to look more benignly upon a diesel that would provide the characteristics they are looking for.

Wrong Answer. Try Again.

According to Freese, in the fall of 1997 he was told that he was to have a running light truck/SUV engine by the third quarter of 1998. "I said it couldn't be done." Chip McClure, president of DDC, and Timothy D. Leuliette, president and COO of Penske Corp., said that was the wrong answer. He was to try again. So on September 27, 1997, a Saturday, the program began. It was code-named DELTA: Diesel Engine Light Truck Application. And 228 days later—which would be 148 "business days" (if the people involved in the project cared about things like regular schedules)—the program was complete. On March 30, 1998, there was the major design freeze. On May 13, they had the first running engine. On August 4, they had the engine running in a vehicle (a Dodge Durango).

The "why?" of the program is fairly straight-forward: DDC wants to have a diesel engine ready for its customers and prospects to have available for tryout and testing in their vehicles. If legislative changes should come to pass in the near-term that have effects on CAFE, then DDC is appropriately positioned to respond to those customer needs.

But the big question is "How did they do it?" The answer to that question should be important to any product development team that's interested in accelerating their time to market.

Here's a breakdown:

1. Decide what needs to be done. This may sound obvious. But too often, product development teams don't have a clear focus on what needs to be accomplished. The technical side of things must be integrated with the marketing side. Freese says that there was a cross-functional (not only including DDC personnel, but suppliers, too) team that was empowered. That means "able to make decisions." They performed planning and analysis, such as quality function deployment (QFD) that permitted them to let customer wants guide engineering undertakings. A word about the supplier involvement: Because this was a fast-track project, there was a recognition at DDC that there are some things companies like Eaton, Bosch, AE Goetze, Arvin, Dana, Uni Boring, and others do exceedingly well. They have core competencies. So efforts were made to leverage them by having the supplier personnel integrated into the DELTA team. But it should be noted that not only did these supplier team members sign on to get the work on the program, but they were involved in issues related to costs and warranties. They were truly integrated into the program with responsibility and accountability. Freese points out that in one instance, DDC personnel identified the characteristics that the valve train required, so they roughed out a design and went to Eaton and asked the people there to do the development. They came back "way over cost," Freese recalls, then adds, "But they did what we roughed out. We asked them to think out of the box"—or to forget about the initial design. "They came back with a design that cost less." That was a clear indication to everyone that suppliers had to do what the suppliers knew how to do best.



2. Provide resources up front. One of the typical approaches to product development in several U.S. companies is to throw resources in as the point of production nears. In other words, start small, then grow big before tapering off. But this is—essentially—a fire-fighting approach. One of the things that Japanese companies are well known for doing is having thorough plans up front so that when they get downstream, their product launches go smoothly and quickly. The DDC DELTA team took a page from this book as they staffed up early.

3. Get everyone on the team on the team. The design targets that were identified in the first step ("Decide what needs to be done.") became widely known. "All of the people involved understood the requirements," Freese observes. Not only was there co-location of team-members, but information was kept up to date in a computer network that the engineers accessed on a local area network (LAN). There were regular reviews: design, analysis, and program reviews. If there were problems that surfaced, they were dealt with, not submerged.

4. Manage the program with program management. Someone recently remarked that too many people think that all they need for program management is a software package. That's far from the case. There must be people who are well versed in using the tool. There must be discipline. Which is precisely what they had on the DELTA project. The critical path schedule was created and tracked on a daily basis. They also tracked the bill of materials that was being developed and its effect on cost. Quality was another concern. One of the things they concentrated on was problem prediction—when going fast, it is exceedingly important to anticipate when things may go wrong so that they can be quickly addressed in the event that they actually do arise. Making all of this happen requires the aforementioned empowered team: a group of people who were both responsible and accountable. And able to take action. "Everyone uses the buzzwords," Freese says. "We didn't say it. We made sure we did it."

5. Do what you plan. Freese describes it as a "textbook approach." There was a structured product development plan created. There were stages identified with deliverables at each stage. So they went through the steps as planned and made sure that the deliverables were delivered when required. When mistakes were made, the response was to fix them right away.

6. Work with incomplete information. This is one thing that engineers have a tough time doing. Freese says that they like to have all of the relevant information before initiating a new step. But in a rapid development program, there cannot be time spent waiting for each sequential step to be brought to completion before the next one begins. So, for example, analysis and design were being performed at the same time (which necessitated the use of computer-aided engineering tools so that there could be rapid modification: run an finite element analysis of a part, see where there are problems, and quickly make changes to the design). The inclusion of manufacturing people in the DELTA program resulted in their being able to actually assemble the first engine in just two days.

7. Use time-leveraging technology. They used rapid prototyping tools that permitted the creation of solid, physical models to verify designs. Not only did they work with 3-Dimensional Services, a rapid prototyping service bureau, but the supplier of the castings for the block, head, and bedplate, CIFUNSA, provided rapid prototyping capabilities that allowed the fast creation of cores for foundry prototypes.

That's how they did it.

According to Freese, no decisions have been made with regard to where the engine will be manufactured. A lot of this depends on what vehicle manufacturer would want to use it in their light truck or SUV. He does say that it could be in production in 2002. Which is in itself quick.

Changes in the Diesel Business

According to Timothy Leuliette, president and COO of Penske Corp. and vice chairman of Detroit Diesel, back in the 1970s, the Detroit Diesel facility in Detroit was producing some 130,000 engines per year with approximately 18,000 people. Today, there are fewer than 2,000 people producing 80,000 engines. Among the reasons for this change are that earlier, they were producing mainly two-cycle engines, which are complicated. Today, the primary product at the plant is the Series 60 engine for heavy-duty on-highway applications. For that engine, there are just six parts produced in the plant (e.g., head, block, pistons), with the rest being provided by suppliers.

Prior to joining Penske, Leuliette was president and CEO of ITT Automotive. Given that perspective, he notes that one of the things that he discovered about the diesel industry in North America, which is certainly different from his previous experience, is that the diesel industry "didn't have Japanese competition." Consequently, he suggests, the changes that the auto industry in North America underwent in order to maintain competitiveness didn't occur in the diesel business.

Now, he says, they are working to design and produce diesel engines around automotive metrics—such things as quality, reliability, consistency, and reduction of complexity.