Automakers are using French-born manufacturing software to improve the machining and assembly of their vehicles. Carmakers use ILOG software to determine the order of building vehicles that will optimize production, maximizing return on investment. A more recent French software entrant in the Detroit area is ILOG, which opened a sales and technical support office in Southfield, Michigan, in May 2000, to serve the US automotive market. Delmia Corp.’s, a French company, labs in Troy, Paris, Montreal, Stuttgart, and Bangalore, India, customize software services to design, simulate, optimize, and control production activities, which account for up to 80 percent of the cost of manufactured goods. Delmia adapted three of its proprietary software tools to form the core software of the V-Comm Project. The Delmia Assembly Module enables users to evaluate alternative sequences of assembly to achieve the optimal lean solution. Toyota engineers working in V-Comm rooms at 20 Toyota locations in Japan, Europe, and North America use the Delmia software to create virtual prototypes that are projected on large screens, and to observe the visual data in three dimensions.
In late July 1701, a handful of canoes led by French soldier and explorer Antoine de la Mothe Cadillac touched on the north shore of Lake St. Clair to establish Fort Pontchartrain and found a fur trading colony named Detroit. Three centuries later, the French are back in the Motor City, this time trading in manufacturing software.
American car companies and their suppliers are using the French products to optimize assembly sequencing, eliminate bottlenecks, and improve machining accuracy, based on solid experience with multinational automotive clients.
Before the arrival of such relative newcomers to the Detroit area as ILOG and Metrologic, another French software company, Delmia Corp., pioneered a beachhead in Troy, Mich. , in 1985. Delmia, a subsidiary of Dassault Systemes of Paris, is an acronym for Digital Enterprise Lean Manufacturing Interactive Applications, lean manufacturing referring to the highly efficient manufacturing practices pioneered by Toyota in the late 1950s to reduce the time-to-market for manufactured goods.
Delmia's labs in Troy, Paris, Montreal, Stuttgart, and Bangalore, India, customize software services to design, simulate, optimize, and control production activities , which account for up to 80 percent of the cost of manufactured goods. Product teams use Delmia software for virtual prototyping and tooling mockups, and to conduct ergonomic analysis of workstations, with the aim of preventing bottlenecks. Delmia products transmit the process plan immediately to the shop floor, whether to human or robot workers, through distributed controls or personal computers.
Delmia was selected as the strategic software supplier for General Motors worldwide in 1998. "We work with a number of first tier automotive suppliers, that is, manufacturers of brake systems and sea ts, as part of GM's Computer Aided Production Engineering software so that they can model their vehicle assembly plants and implement their best practices worldwide," said Bob Brown, a mechanical engineer and Delmia's president.
Making Toyota Leaner
It is fitting that Delmia, which has "lean manufacturing" in its name, was approached by Toyota, the automaker that coined the term, to develop the company's Visual & Virtual Communications, or V-Conm1, project in 1996. Toyota launched V-Comm to assist in cutting its time to-market cycle from 18 months to one year by improving its digital manufacturing operations.
Delmia adapted three of its proprietary software tools to form the core software of the V-Comm Project. These were Envision with the Assembly option , and ERGO option. "Envision allows yo u to model the manufacturing environment in 3-D, including ergonomic analysis of human manikins, and the man-machine interface between worker and production equipment, so that you can find safety hazards and ergonomic situations that would cause stress in the digital environment before you build the tooling," explained Philippe Charles, chief executive officer of Delmia. "You can also make sure the tooling accommodates a wide range of individuals, whatever their size and stature.
The Delmia Assembly Module enables users to evaluate alternative sequences of assembly in order to achieve the optimal lean solution. Brown offered the example of installing an instrument panel in a car. "There are many ways to do it and different fasteners that can be used," he said. "You can try all of them with the design in question to determine the most efficient. Thus, you can design for assembly taking into account all the human factors."
Toyota engineers working in V-Comm rooms at 20 Toyota locations in Japan , Europe, and North America use the Delmia software to create virtual prototypes that are projected on large screens, and to observe the visual data in three dimensions. By virtue of a Web environment, engineers can collaborate with their counterparts in different V-Comm rooms, and communicate with key suppliers, to simulate and validate the feasibility of their products, and to troub1eshoot and solve problems before manufacturing begins.
This has trimmed development time and costs considerably for the Japanese automaker. "We are now developing a new model vehicle that we target to complete within 12 months," said Kousuke Shiramizu, executive managing director at Toyota. "It was not achieved solely by V-ConU11 capabilities; however, it plays a major role in shortening the time frame."
DaimlerChrysler turned to Delmia GmbH in Fellbach, Germany, in 1996 to optimize the assembly of its Mercedes-Benz A-Class economy car at its plant in Rastatt, Germany. The Delmia staff worked closely with DaimlerChrysler's organization and data processing personnel to develop the Rastatter System zur Analysenerstellung und AusTaktung, a.k.a. Rastatt System for Analysis Generation and Balancing, or Rasant for short . The DaimlerChrysler team brought to the table detailed descriptions of the factory's assembly lines. Delmia contributed its Ergofab software module so the new software program could pelorminstant ergonomic analysis. Rasant was designed to be integrated into established software systems.
Rasant generates work sequences on the b as is of method- of- time measurement analysis-this is the standard time for every task in the manufacturing operation and also considers how the sequences may affect the product. The software balances the assembly lines and assesses the ergonomics of workplace work sequences.
In operation, the Rasant tool displays moving work sequences on a production planner's computer screen, and automatically performs an ergonomic assessment of the stations concerned. The software will identify and pinpoint any bottlenecks in the movement of parts, equipment, or personnel. The planner can use the software to solve the problem by adjusting the automated or manual procedures at those workstations.
"This is of particular value to European auto makers who build their vehicles according to specific customer specifications, so it's rare that more than a few of the same type of cars will follow one another on the assembly line," said Peter Schmitt, a mechanical engineer and vice president of marketing and business development at Delmia. This makes it difficult to optimize production throughput by forecasting specific vehicle assembly.
"Rasant gives production planners the ability to detect problems in advance, and graphically edit them to obtain an equalized workload," Schmitt said. By using Rasant, the Rastatt plant was able to balance its production lines so that each worker is busily occupied for an average of 85 to 95 percent of the time, depending on the current order sequence.
Based on its initial success at Rastatt, where Rasant was first used plant-wide in 1998, DaimlerChrysler installed the software tool at its Vario van plant in Vitoria, Spain, and its Hambach two-seater car factory in Ludwigsfelde, Germany. The software tool also optimizes the production of the M-Class sedans assembled in Tuscaloosa, Ala.
Building Three Nissans on Two Lines
A more recent French software entrant in the Detroit area is ILOG, which opened a sales and technical support office in Southfield, Mich., last May to serve the U.S. automotive market.
Computer science graduates from universities, including the Ecole Poly technique in Paris and the Massachusetts Institute of Technology in Cambridge, founded the parent company in Paris in 1987. ILOG stands for intelligence logicielle, approximately, "smart software." The company opened a parallel headquarters in Mountain View, Calif., in 1994, as well as offices in the United Kingdom, Germany, Spain, Japan, and Singapore.
ILOG specializes in developing embedded C++ and Java software components for use in the supply chain, telecommunications, transportation, and financial services. The ILOG software most applicable to manufacturing is its optimization suite of products, including ILOG Solver and ILOG CPlex. These tools are used by automotive manufacturers, including DaimlerChrysler, Ford Motor Co., General Motors Corp., Honda, Michelin, Mitsubishi, Nissan, Peugeot, Renault, Volkswagen, and Volvo, to maximize their users' return on capital, assets, and payroll.
Specifically, carmakers use ILOG software to model the entire production cycle of their vehicles, then optimize any portion of the manufacturing process with constraint and linear programming algorithms.
"Constraint programming can be used to solve detailed scheduling problems," said Greg lmirzian, a computer scientist and technical account manager at ILOG. "For example: In what order does the client build vehicles to optimize their production? If the company has to build 1,000 vehicles during a given period, the build order can be represented with 1,000 variables." Examples of plant constraints Imirzian described were not building five-door models back to back, and distributing vehicles with sunroofs smoothly throughout the day, so that each hour of production has roughly the same number of sunroofs.
"Linear programming algorithms can be a good approach for higher-level scheduling problems. For example, if we are planning several weeks of production, how many of each type of car should we build each day."
Car manufacturers use ILOG optimization software to run their plants on just-in-time principles, by planning and scheduling plant operations far enough in advance to order materials just before they are needed, so they are put swiftly into production, eliminating stockpiled inventory. The software helps balance the thousands of elements that make up a modern automotive assembly line by matching supplies, equipment, and workers to the operations.
The software lets plant management add new operations to lines already covered.
ILOG software can also schedule particular vehicles for assembly. For example, car manufacturers use the software to group their vehicles by color in order to reduce the amount of time operators need to change paints and clean equipment in paint booths.
"A major benefit of ILOG products is that they are designed to be added on to existing software systems, eliminating the need to replace software systems outright," said Denis Sennechael, a computer science engineer and sales director of the lLOG automotive value chain management division. Sennechael said that [LOG's approach to designing its production optimization products is a flexible one.
"We can work with third-party systems integrators and automotive OEMs to customize our optimization products to their specific plants, or with independent software vendors to create standard optimization modules to work with their products," Sennechael explained.
ILOG's recent work with PA Consulting, a systems integration company based in London, enabled Nissan to build a third model car at its Sunderland, England, car factory, without adding a third production line. The Sunderland plant is a showcase facility, producing about 334,000 autos per year, or 38 cars an hour-approximately 70 percent of all the Nissans sold in Europe.
This plant was already touted as the most efficient car factory in Europe in 1998, when it turned out 100 Micra and Primera sedans per worker, or 7,000 vehicles per week. Then, Nissan decided to build 3,000 of its Almera model sedans there each week, too.
The conventional solution of constructing a third production line dedicated to the Almera would have cost in the hundreds of millions of dollars, so Nissan considered building the new car on its two established lines. This required developing a highly accurate scheduling system to enable different model cars to cross over one production line to another without interrupting overall efficiency.
Nissan already used its own software based on plant manager input to plan production at Sunderland, but many variables, such as part availability and tool changeover, meant that the plant adhered to its schedule only three percent of the time. The automaker approached ILOG to improve scheduling accuracy to accommodate the 33 percent increase in production for the Almera without building a third assembly line or major retooling.
"We worked with our professional services organization to help PA Consulting tailor their sequencing system to the Sunderland plant, and fine-tune the sequencing system," Sennechael recalled. "This took about six months, approximately the same amount of time as for Nissan to implement the system at Sunderland."
The sequencing system, which contains ILOG's Solver optimization software, enables cars being built at the Sunderland plant to cross over from assembly lines to paint lines and tooling lines, as needed to improve throughput.
"This means deciding what vehicles Nissan will build on a given day, then deciding in what sequence those vehicles should be built, to speed up production. For example, they may decide to assemble at least three, but no more than five, consecutive five-door cars, and follow that with at least two three-door cars;' explained Imirzian from [LOG.
Since implementing the software system in the spring and summer of last year, the Sunderland plant meets its vehicle scheduling targets 85 percent of the time. Plant management uses the scheduling tool to study the potential effect of operational or constraint changes before they are made, and no longer needs to reschedule vehicles while they are in the storage buffers between major sections of the plant.
Except for some storage and secondary tooling for the Almera, all three vehicles are made on the two Sunderland lines. The $2 million cost of the software system was paid back within three days of production.
Daimlerchrysler Beats the Clock
An example of how ILOG can work directly with automotive OEMs involves DaimlerChrysler, which was able to begin production of its Chrysler PT Cruiser on June 30- two weeks earlier than expected- at its Toluca, Mexico, factory by using its Centralized Vehicle Scheduler system, which relies on [LOG's Solver optimization software. The CVS system was recently deployed on the vehicle manufacturing and painting lines of the PT Cruiser and enabled an increase in annual production at DaimlerChrysler's Toluca plant by 4,000 vehicles without additional tooling.
This is helping DaimlerChrysler meet the growing demand for the popular, retro-style sedan. More than 45 ,000 of the popular PT Cruisers have been sold since the model was introduced at the 1999 North American International Car show.
The CVS system is run on a standard personal computer and a Sun Microsystems Solaris platform, replacing previous scheduling applications that ran on mainframe computers. The newer scheduling system downloads scheduling data from company servers, processes it, and returns optimized data to the PC within minutes, rather than the full day it took previously.
" lLOG optimization technology not only solves problems that were once impossible, it does the job quickly. This is a tremendous advantage because in this industry, time isn't just money; it's survival," said James Whitfield, manager of centralized vehicle scheduling and forecasting.
DaimlerChrysler is using the [LOG-based CVS system at its headquarters in Auburn Hills, Mich. , to schedule production at its 18 North American and European assembly plants. The aim is to optimize the production of all Chrysler, Jeep, and Plymouth vehicles in the United States . DaimlerChrysler expects the greater scheduling efficiency of the CVS to save approximately $7 million in painting operations, and another $10 million to $20 million for machining and assembly tasks.
Closing the Digital Loop
Metrologic Group Services Ine. of Farmington Hills, Mich ., was formed in 1999. Although it's new to the States, the developer of measurement software draws upon 20 years of experience earned by its parent company, Metrologic Group S.A. of Meylon, France. Two 19-yearold electronics engineers named Philippe Cimadomo and Gilles Bartoli created the Metrologic Group in 1980 to design hardware products to equip measuring instruments and machinery. By 1985, the duo began developing their own measuring and control software for coordinate measuring machines, or CMMs, that were becoming popular in mass part production.
CMMs are equipped with sensors that inspect the shape and dimension of manufactured parts and software that can read CAD data. "The most typical application is comparing the finished part with the CAD drawing to see what you've actually made to ensure you are meeting the design intent," said Bryn Edwards, chief executive officer of Metrologic Group Services.
Metrologic used its own software to design a suite of hardware to retrofit older CMMs and improve their efficiency. This hardware includes counting boards to serve manually operated CMMs, digital readouts to enable part manufacturers to view CMM coordinates and impose them on parts they are machining, and computer numerical controllers to direct CNC tooling accordingly.
A study conducted for Metrologic by Ing Barings of Amsterdam and London reported that automakers account for 20 percent of the approximately 8,500 CMMs that are installed worldwide each year. However, carmakers also drive use of the machines by their component suppliers, because CMM enables problems to be rectified during the manufacturing process rather than at the end of the cycle. Ing Barings concluded that, overall, the automotive industry accounts for 80 percent of all CMM business.
This reasoning was not lost on the Metrologic Group. In 1996, the French company signed an agreement with Volvo to retrofit and equip all the Swedish carmakers' CMMs with software and hardware. Two years later, Peugeot, Renault, and aerospace firm Aerospatial agreed to retrofit their measuring machines with Metrologic Group's products.
Metrologic plans to use its latest measuring software product, the Metrologic IT, to crack the U.S. automotive CMM market. Metrologic designed this 3-D tool to be installed with new CMMs, or to retrofit older machines, whether manual, robotic, touch sensor, laser, or video. Metrologic claims Delmia's Rasant software balances the assembly lines and assesses the ergonomics of workplace labor sequences, so that workers are busily occupied between 85 and 95 percent of the time. its software is compatible with virtually all CMM models, including Brown & Sharp, Zeiss, LK, and DEA touch sensor CMMs, Leica, SMX, and API laser trackers, and Faro and Roma articulated arm machines.
CMM operators program the Metrologic II to measure pertinent defined geometrical features of a part and to generate the automatic probe paths to measure that part. The software makes its calculations using algorithms certified by the National Institute of Standards and Technology in Gaithersburg, Md., and Physik Technologie in Braunschweig, Germany.
The software creates a detailed graphical display of every geometrical feature, including probed points, and graphical display of any deviation from the CAD design . Operators can select a variety of geometrical constructions for those displays to suit the part.
In addition, Metrologic II can be aligned as needed; for example, 3-2-1 configuration for position; that is, taking readings from three points atop the part, two from its edge, and one on another point to lock the part into the six degrees of freedom for a machine tool. Operators can also use a reference feature, such as a facing or a hole, as datum to configure the software.
Edwards described Metrologic II as a feedback system that enables part manufacturers to improve their machining process. "These days, all design work for parts is done digitally on a computer, without the construction of models or artifacts," he said. "CNC machines then manufacture the part digitally. Although the design intent is perfect-as a CAD drawing-the actual processing of the part is not, because even with CNC machines, there is always an uncertainty, from human error to manufacturing uncertainty, such as cutting controls, stamping process stretch, and metal flow characteristics. Metrologic II enables parts makers to measure their parts digitally and compare them to the original design intent, thereby closing the loop between digital design and manufacture."
Beyond improving the tooling of standard parts, CMM operators can use Metrologic II to evaluate the shape of a new part, or to improve the recalibration of machining fixtures. "They could also use the software to study a competitor's part, and to reproduce it, whether car parts, castings, or turbine blades," remarked Edwards
Metrologic has begun supplying hardware, some of it equipped with Metrologic II, to U.S. automakers, including BMW in Greensville, S.C.