Partnerships, simulations, and better information control help automakers turn a concept into a new car in just two years.
Automotive Segment Manager
Getting a new car onto the production line and into the showroom in the shortest possible time is critical to the economic survival of the world’s automakers. New car programs that took four years not long ago now take a mere two. And that’s just a goal. When they hit that bogey, auto planners will start thinking in terms of months, not years.
The need for speed relates to the shifting fancy of the consumer. Remember when Chrysler introduced the minivan? It was such an unexpected hit other automakers were left scrambling to introduce their own versions. Sport-utility vehicles have been the latest rage, and now that just about every car company offers one, it’s on to the next hot segment, whatever it will be.
The two-year development period represents the time it takes an automaker to bring a well-defined concept into production. They spend at least a year before that conducting market research on the niche they’re trying to fill, and evaluating several styling themes and functional mock-ups. The two-year period is also when CAD, CAM, CAE, and PIM (project information management) tools have the most impact. The tools played a big role in chopping the previous fouryear development period in half. And they’re key to future improvements.
24 MONTHS, THEN 23, THEN 22...
To make the best use of CAD, CAM, CAE, and PIM technology, or C3P in the terminology of Ford Motor Co., automakers are focusing on several proven strategies.
One avoids data translation whenever possible. Reformatting design data for use in another program always calls for a certain amount of “clean up.” Rather than waste time, Ford engineers declared that suppliers could no longer send IGES data, a frequent source of flawed models.
Ford and its suppliers instead rely on over 2,000 seats of SDRC’s I-DEAS solid-modeling Master Series software for the 17 automobile programs underway. The automaker and its suppliers communicate directly through solid models that move from company to company without translation from their original format. Avoiding data translations also allows earlier use of powerful capabilities such as computer simulations.
When simulation results aren’t available in a timely manner, they get cut from the design process. This happens when models must be translated or rebuilt to work in different software. Companies lose the ability to uncover problems early, when they are easier and less costly to fix. Problems that might have been detected through simulation don’t show up until prototype testing, and then they can delay a project for weeks.
For Dortec Industries in Newmarket, Ont., a Tier-one supplier of electronic door-closure systems to Ford, one main requirement for its CAD system is the ability to use design data in its original format for analysis. “Our industry reached the point where we had to reduce the art-to-part cycle,” says John Stokes, MIS manager at Dortec. “We didn’t have time to translate from CAD to a preprocessor, and from the postprocessor back to CAD.”
Similarly, Navistar’s Engine and Foundry Div., a supplier of diesel engines for Ford trucks, plans to cut cycle time by 50%. They say they’ll reach the figure by avoiding data translations and conducting more up-front computer simulations. “Our use of analytical methods was once limited to postmortem modeling,” says Dennis Jadin, the division’s chief engineer. “Not enough 3D geometry for early analysis resulted in an inefficient and lengthy model building process, and that diminished the impact of analysis in our product development process.” Jadin’s problem was that 3D geometry for analysis was created from 2D drawings, and usually after a part was released to production. “Now we get the design geometry and let engineers run an analysis or send it to the Design Analysis Group,” he adds. “This eliminates discrepancies and redundancies in models and streamlines the product development process.”
Jadin’s experience also shows how issues intertwine and that solving one almost solves another. For instance, once data translations are eliminated, more designs are available for simulations.
A simple cowl box at the base of a windshield is a case in point. It directs air to the air conditioning and heating systems. Water rolling off the windshield entering the box should exit through drains at opposite ends of the cowl. Physical testing showed that a blower also forced water into the airhandling system, even at low vehicle speeds.
Because airflow through the box is too complex to visualize with physical testing, Ford engineers turned to computational fluid dynamics (CFD) to improve the design. The analysis showed that an air velocity high enough in several areas on the outboard cowl could pull droplets into the inlet. Vivid images of results made it easy to understand the problem. Engineers added a hole to reduce air speed and a cover to deflect water from the air inlet, solving the water-ingestion problem.
Simulation efforts range from component- level stress and CFD analyses to reviews of entire assembly processes. Manufacturing simulation, for example, helps Ford design its vehicle-final-assembly processes faster and at lower cost. The automaker added the “virtual factory” concept to its C3P strategy that lets manufacturing engineers work alongside product design engineers, both using the same computer images of vehicle components and facilities. Design changes are made immediately and early enough to improve a component’s easeof- assembly into a vehicle.
Virtual-factory techniques have already been used to design a door module, fuel-tank placement, door latch and drive-line assembly on cars, and a light truck’s instrument panel. Ford estimates that virtual-factory concepts can cut 20% off manufacturing-driven design changes when launching a new vehicle.
BETTER DATA MANAGEMENT
Most automakers and their suppliers realize that CAD, CAM, and CAE tools are underutilized when not accompanied by some sort of data-management capability. Ford has made no secret that its decision to standardize on one modeling system was based in part on the availability of Metaphase Enterprise, a PIM system from SDRC.
Other companies have also recognized the importance of PIM. As engineering departments were transformed by computers and design-automation technology, a not-so-positive result became the inability to track and control product information. Different versions of design and engineering data multiplied. Incompatible databases proliferated. Part-classification schemes grew rife with redundancies, and outdated and inaccurate information, making data management essential.
While some refer to this function as product-data management, Ford has chosen the term Product Information Management in recognition of the importance of providing information instead of just data. The PIM selected by the automaker ensures that people will reliably and consistently base decisions on current information, and have confidence in the accuracy and timeliness of the content.
The information management system relies on the company’s corporate intranet for its technical architecture. The system also connects to other non-Web-based infrastructures. The intranet and Metaphase provide a single point of contact for the company’s knowledge, acting as a library of information on product, process, and lessons learned.
Ford and its Tier-one suppliers are cooperating by acquiring the same PIM systems to share information. Implementing the virtual-corporation idea means everyone involved in the development of an automobile will have access to product data along with the intelligence that went into it, regardless of their physical location, application, or operating system. The information will move with the product as it goes from design to manufacture to maintenance.
The intelligence that gives each component meaning will be automatically maintained by the PIM system, along with the relationships among product data elements, product history, configuration-management information, and other essential elements in the entire product life cycle. Such information comes from a single source and must be accessible to various users in a way that supports their job. It’s an ambitious goal, but Ford believes it’s key to slashing cycle time.
Although not related directly to the use of technology, another way automobile companies are cutting cycle time is by converting firms that were formerly suppliers into partners. That can mean different things, depending on what the supplier delivers. A component supplier, for example, takes on more design responsibility. Rather than simply delivering a part to the OEM’s specifications, a supplierpartner receives basic parameters and then designs the part. Spreading design tasks over more resources cuts cycle time. Becoming a partner also means bearing more financial responsibility for components produced. The latest wrinkle is giving some suppliers sleepless nights because they will now share warranty costs with the OEM for their products.
Auto companies are also expecting their software provider to become partners. Ford, for one, selected a CAD vendor that would act as a partner in their C3P effort. A close working relationship with the vendor was more important to the auto maker than the software’s technical specifications.
For its part, SDRC has about 300 people working with Ford to help deploy the C3P system. SDRC is helping implement IDEAS Master Series and Metaphase on new vehicle programs and training new users to maximize productivity. Being a partner to the auto company involves other activities such as helping the automaker get its suppliers to use I-DEAS Master Series through financial incentives and special programs, and creating proprietary customizations of the software for Ford’s use.
Seamless data transfer, extensive simulation, better use of information, and partners — these are ingredients of the auto industry’s strategy for shrinking time to market. They’ll also work for any industry that requires the collaboration of many different groups to get a product to market.