PLM: Boeing's Dream, Airbus' Nightmare
- An Urgent Message
- Rewards, and Risks, of PLM
- Airbus' Nightmare: Born in a Storm
- Boeing Takes a Different Flight Path
- Dreamliner's "Virtual Takeoff"
Also in This Feature:
- Top Players Inside and Outside Boeing
- Technologies To Help the 787 Take Flight
- Project Planner: Calculating Costs of a Product Life-Cycle Management System
- By the Numbers: Boeing and Airbus Battle for Air Supremacy
Question: What's the best way to prevent software incompatibility problems? Write a letter to the editor at: firstname.lastname@example.org
An Urgent Message
For Kevin Fowler, Tuesday, Oct. 3, 2006, was another long day at Boeing. As the vice president of systems integration, processes and tools for the 787 Dreamliner, the company's first completely new airliner since the early 1990s, it's his job to ensure that all the pieces come together smoothly. As part of those duties, he was leaving the company's offices in Everett, Wash., to catch a flight to France, where he would pay a visit to Labinal, the wiring supplier for the 787. On the way to the airport, an urgent e-mail landed in his BlackBerry.
The message was from his boss, 787 program chief Mike Bair, and he wanted to know if Fowler had heard the news out of Europe.
Half a world away, Airbus chief executive Christian Streiff had delivered a speech announcing that the company's A380 superjumbo would be delayed by at least two years. The delay and resulting changes to the program were expected to cost Boeing's fiercest competitor as much as $6 billion in lost profits, by Airbus' own reckoning. The cause, Streiff said, was due to compatibility issues with the sophisticated computer-aided design software used by engineers to architect the A380.
Airbus engineers in Germany, where the plane's rear fuselage section was being built along with the hundreds of miles of electrical wiring that power the main cabin, were using an older version of Dassault Systèmes' trademark Catia computer-aided design software—version 4. Engineers in Toulouse, France, where the A380s were being assembled, were using a newer version of the software, Catia V5.
When the first wiring bundles, large packs of preconfigured wires to power everything from lights to in-seat entertainment systems, began arriving at the assembly plant in Toulouse last June, Streiff said they didn't fit properly from the rear section into the front section of the fuselage. Workers tried to pull the bundles apart and feed the wiring through the fuselage by hand, but with 300 miles of wire and some 40,300 connectors on each plane, the immensity of the problem soon became obvious. An unthinkable blunder had happened—as the computer-aided design files were passed between the different versions of the Catia software, the company said errors occurred. And software experts familiar with the incident say the errors included changes in measurements. Those errors are going to cost Airbus billions.
"The root cause of the issue is that there were incompatibilities in the development of the concurrent engineering tools to be used for the design of the electrical harnesses installation," Streiff said in his October speech about the delay. In other words, information from the two systems could not be exchanged accurately, or at the very least, Airbus did not integrate them properly. "We have to update and harmonize the 3D-design tools and database—and it will take time to do this," he added.
The question in the e-mail on Fowler's BlackBerry back in Everett was blunt. Boeing was using the very same Catia software from Dassault as a cornerstone of the 787 program. Could the same errors derail the Dreamliner?
"My answer was no," Fowler says. "I have confidence in our program. I think we understand what it takes to put together the product, and we understand what the issues are. I also believe we've taken the right steps to prevent the same thing from happening here."
But others aren't so sure. Computer-aided design (CAD) software, such as Catia, often anchors product life-cycle management (PLM) suites, software packages that have grown in popularity as well as in sophistication and complexity, along with computing horsepower and collaboration technologies like the Internet. Today, they are capable of managing every aspect of a product's life cycle, from concept and design, to manufacturing, maintenance once the product is sold, and even through to its eventual retirement. In effect, they can follow a product, be it a toaster or jet, from cradle to grave.
Like any other large software implementation, be it enterprise resource planning (ERP) or customer relationship management (CRM) systems, there is plenty of room for error. "On paper, PLM looks like it delivers a perfect world," says Robert Bean, chief operating officer of Kubotek USA, a Marlborough, Mass., firm that specializes in CAD systems. "But we're not living in a perfect world. And when you're talking about something as complicated as an aircraft, you're dealing with a massive movement of data." The data warehouse for the Dreamliner project, for example, is 16 terabytes.
In fact, Kubotek released a study in October, the same month Streiff revealed the extent of the problems at Airbus, highlighting widespread compatibility problems between different CAD programs. The survey, which queried 2,800 CAD engineers, found that 50% of all respondents indicated they had to redesign a new part or tool on a weekly or more frequent basis after 3D models were exchanged between CAD systems, due to errors that were introduced such as changes to measurements.
For now, Airbus' woes have provided a boost to Boeing and its Dreamliner program. Boeing finished 2006 with its best year ever for commercial aircraft sales, taking in firm orders for 1,044 planes and beating Airbus for the first time since 2000. And with the Dreamliner raking in 157 firm orders in 2006—making it Boeing's most successful new launch—Fowler and the 787 team have every right to be feeling the wind beneath their wings. Still, say industry observers, they would be wise to keep their fingers crossed.
"If it happened once, it could happen again," warns Ian Massey, a former controller with Airbus, and now an executive with airplane financing company Republic Financial of Aurora, Colo.
Reward Versus Risk
The missteps at Airbus, and Boeing's own reliance on Dassault's sophisticated software, underlie the stark reward-versus-risk scenario offered by PLM suites.
PLM packages, like CRM and ERP systems, consist of a set of integrated programs designed to automate certain processes within a business, be it customer service or finance. In the case of PLM, the software is focused on the development and production of products. The number of modules integrated into a PLM suite varies from vendor to vendor and the industry being addressed, but there are usually three core offerings: a CAD system like Catia; a digital manufacturing system, such as Dassault's Delmia, which allows companies to simulate how a product will be manufactured; and a product data management (PDM) system, such as Dassault's Enovia, which manages all the data associated with a product, such as CAD drawings and specifications.
Using the CAD software, manufacturers can create detailed 3D models of their products and run those designs through a battery of virtual tests, such as stress, vibration, noise, wind and even crash tests, long before a cent is spent on manufacturing. Using the digital manufacturing software, companies can explore how those parts or components can be produced by simulating the process. Products like Dassault's Delmia help manufacturers determine how many people, robots or other manufacturing resources will be required, whether existing machinery can be utilized or new purchases will need to be made, and whether processes can be automated or will need to be performed manually.
While the PDM piece of the suites is focused on managing the data associated with a product's development, it really forms the basis of a collaboration platform. Using software such as Enovia, companies like Boeing can hand off responsibility for designing parts or components to those partners who will ultimately be responsible for their manufacture. Designs can be polished, and the most efficient and cost-effective methods of production can be determined long before the first length of sheet metal is stamped.
The aerospace industry is far from alone in latching on to the potential benefits offered by PLM. Most Fortune 500 manufacturers, from Toyota (which uses Dassault's complete PLM suite—Catia, Delmia and Enovia) to General Motors (a UGS client), golf club manufacturer Ping (Parametric Technology) and consumer goods makers such as Playtex (Agile Software), are at various stages of implementing PLM suites.
Potential cost savings vary depending on the complexity of the product being manufactured, but some analysts, like AMR Research's Michael Burkett, estimate that the time and cost savings on more complicated products can be as much as 50%. That has translated into PLM becoming a hot software category, growing at an annual compound rate of about 8.3%, according to research firm CIMdata of Ann Arbor, Mich. In a report released in October, CIMdata forecast that investments in PLM would grow from about $19 billion in 2006 to $27 billion by 2010.
But underlying that promise are very real dangers. By relying so much on software, and by eschewing real world mock-ups or trials in favor of virtual models and tests, companies may not catch errors until late in the manufacturing process. And the potential pitfalls, such as compatibility problems between different CAD packages, are more common than most chief information officers might think, according to Kubotek COO Bean. Other challenges include everything from maintaining a current and accurate data warehouse for product information, to ensuring that multiple manufacturing partners have the latest software updates, dealing with user training issues and gaining executive support so that rules can be enforced across department and inter-company boundaries.
The risk is compounded by the fact that companies are basing their product rollouts, and by extension their very business, on the platforms. Mistakes can exact a heavy price.
In 2003, the British National Audit Office, which scrutinizes public spending on behalf of Britain's Parliament, pointed to difficulties in implementing CAD software used on the country's Astute nuclear submarine project as a chief source of the program's $1.7 billion in cost overruns. The chief contractor on the project, BAE Systems, said it underestimated the complexity of training users and the amount of information that the 3D design software would need to manage on the project. BAE was using Cadds5 software from Parametric Technology of Needham, Mass. The Astute program is now four years behind schedule, according to the audit office.
Parametric acknowledged that the Astute program did encounter problems related to the implementation of its software, in particular the Cadds5 3D design software. However, spokeswoman Nicole Rowe says that by working closely with the customer, the program has managed to get back on track. "Oftentimes, especially when you're dealing with something as complex as a nuclear submarine, there's a steep learning curve," she says. The first submarine is scheduled to launch on June 8, about seven weeks ahead of the revised schedule.
In addition to the problem encountered with wiring on the A380, Airbus also suffered another setback in February 2006, when a wing on the giant 550-seat aircraft snapped during a stress test. Software modules integrated with CAD systems are meant to simulate stress and catch possible failure points before components go into production. But Airbus said that during the last step of testing in February, where the wingtip is bent more than 24 feet away from its normal position, a rupture occurred between two engines. It did not provide further information on why the rupture could not be prevented through stress simulations; however, Airbus said the rupture would not result in major delays and could be corrected with "refinements" to the design.
Baseline made repeated attempts to speak with Airbus for this story, contacting officials in Europe and North America, but Airbus declined the requests.
Airbus' $6 billion problem offers technology leaders and chief executives a number of stark lessons in what can go wrong in implementing sophisticated PLM systems. But even more important, Boeing's response to the problems at Airbus, and its own use of PLM on the 787 Dreamliner program, offer guidance on what can and should be done to avoid the same turbulence.
Born in a Storm
Airbus' 2006 nightmare with PLM can actually be traced back to the giant company's difficult birthing process in 2001. "This issue dates back to the historical structure of Airbus," recalls former Airbus financial executive Massey. A loose consortium of French, German, British and Spanish companies formally spun out Airbus in 2001 at the same time the A380 program was being launched.
Massey well remembers the infighting among the partners over jobs, and over which country would get the bulk of the mammoth aircraft's production work. Some executives, in fact, expressed the feeling that, as Massey puts it, "We shouldn't be launching this aircraft put together by four different nations." Such disputes can have a downside, resulting in a level of distrust or, at best, erratic coordination. "Because there was an awful lot of debate about the way to create this single-company structure, the A380 was held hostage to that," he says.
For example, the Germans were adamant that the entire aircraft not be built in France, where Airbus was headquartered. Ultimately, work on the A380 was carved up among the four players, so that at its founding in 2001, Airbus had offices and factories at 16 sites spread across four countries and employing 41,000 people. Each country had a level of independence to go its own way when it came to systems and technology, Massey points out.
This lack of strict uniformity of processes and technologies laid the seeds for what was later to grow into an entangled vine of trouble for Airbus. "The systems had been set up under the old structure," Massey says. "No one was watching who was using what versions of Catia. It may be a systems issue, but as much as anything, it's a management issue."
Airbus' lax enforcement of a single lingua franca for design was at the heart of the A380's later problems. While there are many ways that different CAD systems, and even different editions of the same CAD programs, can trip up a product's design, those ways become multiplied with the complexity of the end product and the increased number of suppliers creating parts or components for its manufacture.
By contrast, Boeing management is taking no such chances. Well before Airbus' problem became public, the U.S. aerospace manufacturer had put into place a rigorous set of requirements to ensure that the same edition of Catia is used by everyone connected with the shaping of the Dreamliner.
At least one Airbus design manager was well aware of the potential for a CAD incompatibility disaster. Martin Horwood, lead engineer for CAD capability development at Airbus U.K., co-authored an article titled "CAD Data Quality" in the May-June 2005 issue of Engineering Designer magazine in which he warned, "With data arriving into the digital mock-up from a globally dispersed design community, including industrial partners, suppliers and subcontractors, it is imperative that the CAD data is of the right quality. Failure … will cause the digital mock-up to be inaccurate and not fulfill its task, leading to expensive reworks in real life."
And fail it did. With its German designers creating wiring bundles to fit inside one set of spaces in the A380's fuselage using Catia V4, and the French designers having created the fuselage wiring spaces using the more modern Catia V5, the actual wiring bundles were unable to fit.
Says Peter Schmitt, vice president of marketing and communications at Dassault Systèmes of America, "The 6 billion [dollar] loss at Airbus was the result of a fairly simple problem that could have been fixed with a fairly low investment." Schmitt didn't offer an estimate of how much it would have cost or how long it would have taken for Airbus to upgrade the German unit to Catia V5. But his message was clear: Companies using PLM should make sure they are using the same software package and version of that software. "Manufacturers using PLM," Schmitt adds, "should make sure everybody is working with the same set of data."
Although Airbus has remained mum on exactly why the German designers used an older CAD package, most observers believe the reason was simple Eurodollars and Eurocents.
The cost to train the engineers in Catia V5 may have been the sticking point for Airbus management that led to the A380's multibillion-euro design flaw. That's the view of an executive at a firm that trains Airbus' suppliers to use Catia. "Airbus made the decision not to migrate Germany to Catia V5 because it would have meant a complete retraining," says Geoff Haines, managing director of Cenit Ltd. in Oxford, England. "They decided not to do it for budgetary reasons."
So great is the chasm between the two versions that someone schooled in Catia V4 trying to get up on V5 is similar to a motorist learning to fly an airplane. It takes six months to a year before they become fully proficient, Haines says. "It would be like starting from scratch," he adds.
Those unfamiliar with CAD software may be wondering just how two versions of the same software package could be incompatible, or for that matter, require such extensive retraining. The reason is the two software editions differ in their basic treatment of drawings, so the way digital models are created is different.
Both systems are able to represent objects in 3D, but that's where the similarities end. Engineers using Catia V4 must use a manual process to create the geometry of a model. To create a hole inside an object, for example, the system requires them to subtract a cylinder from the space to define where the hole should exist. By contrast, the product designer using Catia V5 simply feeds in a set of engineering instructions—in effect, describing the location and dimensions of the hole—and the geometry is automatically created. "V5 is higher-level, more intuitive," says Doug Cheney, product manager for CAD interoperability quality at ITI TranscenData, a developer of CAD translation software. "With the older system, the engineer figures out the geometry; with the new one, the system finds the best geometric solution."
Airbus engineers ran afoul of this basic difference when creating the miles of wiring to be inserted inside the A380 fuselage. The engineers' "notes"—appendices that describe details of models such as curves—sometimes are not replicated in the translation between Catia V4 and V5, says David Prawel, president of Longview Advisors, a Loveland, Colo.-based consulting firm specializing in 3D software issues for manufacturing. In other words, key notes required to duplicate a 3D model showing electrical wires as they twist and bend through the aircraft may fail to reappear in full and accurate detail when a design file in one system is converted to a file in the other.
For example, something basic such as the tolerance level of a metal part, noted by an engineer in the appendix to a 3D drawing, may be left out when the model is converted from one system to another. The result can be that the manufacturer—or a supplier—may produce the part to the wrong tolerance.
In addition, units of measurement, when carried over from one CAD system to another, can create havoc for designers, says Brian Barsamian, president of V5 Engineering, a Newport Beach, Calif., firm. Barsamian trains engineers—including many designers at Boeing—to use Catia V5. "There was a complete rewrite of the code from V4 to V5," he says. "You have to be careful to set parameters defining whether you are exporting metric units or English units; otherwise, a 1 millimeter part can become 25.4 millimeters, because it sees 1 millimeter as 1 inch."
Still, most CAD vendors offer their customers a smooth path to convert their data from an earlier version to a new one, according to Prawel. "Every vendor does a good job of backward compatibility except Dassault," he says. "Why some of the biggest aerospace companies and automotive manufacturers in the world didn't force them to do a better job of backward compatibility is a mystery to me. Now Airbus is paying the price." To solve this problem, Prawel says many Dassault customers have decided to start from scratch to re-create, or remaster, the data in all their existing models in Catia V5 because of its lack of smooth interoperability with the earlier version.
Dassault owns up to the programs' dissimilarities, and the potential minefield they pose for manufacturers adopting PLM midway into an upgrade cycle, or just trying to get the pair to coexist. "If in one organization they are using both versions in parallel and have to synchronize data on a constant basis, that is what causes problems," Schmitt says. However, he says using the two versions does not necessarily spell trouble. Schmitt notes that Dassault has several customers that have successfully used Catia V4 and V5 concurrently on long-term projects.
Another area where Airbus tripped up was in the 3D digital mock-up of the A380. Both companies, Airbus and Boeing, use a digital mock-up as a final design step. Boeing is meticulous in its 3D mock-up process, having had extensive experience in using a virtual mock-up for its earlier 777 jet and in later iterations of its top-selling 737 plane.
The 3D digital mock-up of the A380, however, was done well behind schedule, with a new design team that was under pressure to get it completed. "The problem is the fact that the 3D digital mock-up, which facilitates the design of the electrical harnesses' installation, was implemented late, and that the people working on it were in their learning curve," states an Oct. 3, 2006, Airbus press release. The company signed its first major contract for mock-up software just this past year.
Different Flight Path
When the Boeing board of directors gave formal approval to the 787 Dreamliner program in April 2004, work had already been going on for two years behind the scenes to get ready for the formal launch. In early 2002, Carol Pittman, then information-technology director for the 787 program, began meeting with Fowler, the systems integration chief, to sketch out a technology strategy.
Pittman and Fowler agreed to anchor the Dreamliner on Dassault's PLM platform, largely based on the success Boeing achieved in the all-digital design of the 777. They would use all three major components of the Dassault PLM suite: Catia V5 for the design; Delmia, the virtual manufacturing package that allows partners to take the electronic designs created in Catia and simulate how those parts or components will be manufactured on the factory floor; and Enovia, the collaboration platform that provides engineers with access to the master vault of information on the 787, such as electronic designs and component specifications.
One of the first critical strategies they agreed upon was ensuring software compatibility. Fowler says Boeing was already well aware of the difficulties that could be encountered from exchanging information between different CAD systems through Boeing's extensive experience working with various CAD packages throughout its operations.
Fowler says he wanted to avoid such trouble on the Dreamliner. There was simply too much at stake.
While it has recently been stealing business from Airbus as airlines look for alternatives to the superjumbo, that hasn't been the general trend. In fact, Airbus overtook Boeing in 2001 as the world's largest manufacturer of commercial airplanes and led every year until Boeing's comeback in 2006. Aside from competitive reasons, the Dreamliner also represents a dramatic change in manufacturing for Boeing—one that makes ensuring software compatibility that much more critical. When the Dreamliner takes to the skies in 2008, it will be the first commercial jet to have a fuselage and wings made almost entirely of plastic-like composite materials—mixtures of high-strength fibers, resins and carbon. The new materials, combined with other advancements such as an improved aerodynamic design and more efficient engines, will allow the 787 to burn 20% less fuel than comparable airliners and achieve a 10% to 20% reduction in maintenance costs. The project's development budget is confidential, but estimates are in the range of $8 billion to $10 billion.
In the past, the standard practice for Boeing has been to design the plane in-house, then pass blueprints for parts or whole sections of the plane to manufacturing partners. This time, however, Boeing is turning that process on its head, designing the 787 in collaboration with its partners using the PLM software from Dassault. Essentially, some 6,000 engineers around the world are jointly designing and engineering the aircraft. Partners include companies such as Alenia Aeronautica of Italy, which is building the plane's main fuselage; Japan's Kawasaki Heavy Industries, which is also building part of the fuselage as well as the wings and landing gear; and Goodrich Aerostructures of Chula Vista, Calif., which is constructing the nacelles (shell around the engines) and thrust reversers.
"There are a number of advantages to putting the people closest to the work in charge," Fowler says. The manufacturer of the fuselage, for example, will ultimately know the most cost-efficient method to build the structure. Component manufacturers can point out whether their existing machinery can manufacture a part, or whether new robots or tools will need to be purchased. By altering the design, say, by using a 6 millimeter fastener instead of an 8 millimeter fastener, they may be able to produce the part with existing machinery or manufacture the part faster, saving time and money.
Pittman and Fowler agreed that all engineers working on the 787 would work in Catia V5—no substitutions. This is not as simple as it sounds. For starters, it requires a large up-front investment. Boeing and its suppliers are paying an estimated $20,000 per desktop for the software, which, based on 6,000 engineers worldwide, works out to about $120 million. In addition, engineers do not always adapt well to being told what software to use. Most have spent years learning how to use a specific software package, often customizing it to meet their preferences and learning through experience exactly how digital designs translate into actual engineering.
"We considered allowing our partners to use their own preferred applications, but we decided that wasn't feasible because of the [data] integration challenges," Pittman explains. "It wasn't a popular decision, and we really had to work on explaining why we were doing it."
Boeing provided its suppliers with a financial incentive to get on board with Catia V5. "If you use the common Catia tool, Boeing will provide you with the tool and the support for free," says Barsamian, who trains Boeing engineers to use the software.
Another key plank in the company's strategy was ensuring software version control. Even though all Boeing engineers and partners were starting off with the same version of the various software packages, there is ample opportunity to lose control as updates are released and new partners are brought on board. The team decided that software updates would take place at four specified points in a year—referred to as Block Points—and that all Boeing engineers working on the Dreamliner and all outside partners would receive software updates at the same time.
Again, this understates the complexity of the task. For starters, the updates include far more than Dassault's software; they involve dozens of other applications that are used in the design and engineering process to do everything from test the stress tolerance of composite materials to achieve optimum aerodynamics. Many of the applications have been internally developed by Boeing; however, a number have been developed by third-party vendors, such as Metrologic, whose software is being used for analyzing 3D measurements. In all, some 150 applications are updated at each Block Point.
The updates also include software from other PLM vendors. Boeing is using Windchill, a software package from Parametric Technology in conjunction with Dassault¹s Enovia, to streamline the process of managing changes to components on the Dreamliner. If changes are made to a window design, for example, those changes need to be conveyed to manufacturing partners and internal Boeing designers working on areas affected by the change. Parametric¹s Windchill manages that process, ensuring that engineers follow a consistent set of steps to resolve any conflicts and that changes are completed as requested. Brian Shepherd, vice president of product management for Parametric, notes that even when companies try to consolidate on one vendor¹s PLM offering, they often find certain functions, such as change management, are still best handled by third-party software.
In all, some 150 applications are updated by Boeing at each Block Point on the Dreamliner program.
A final cornerstone of the Dreamliner technology strategy involves the use of a master data repository for all design and engineering information. Enovia, the Dassault platform, is used as a gateway to a 16-terabyte data warehouse in Bellevue, Wash. Boeing encourages its partners to send updates to the data warehouse at least twice a week, and sometimes more frequently depending on the stage of work in progress. The warehouse is housed on Unix servers running IBM's DB2.
Boeing chose to use CAD and PDM systems from the same software firm, Dassault, thereby ensuring tight integration. Airbus, on the other hand, decided to mix and match. The European aerospace company is using a CAD package from Dassault and a data management system from Parametric Technology. In September 2005, Parametric announced that Airbus was extending the use of its data management solution, Windchill, as the platform for managing all CAD models for the A380 that are used in its digital mock-up.
But mixing and matching your CAD and data management vendors can require extra work to ensure a smooth fit. "If you want deep integration with your CAD data, it's best to go with a PDM system from your CAD vendor," Cheney says.
There are signs, though, that Airbus is already having second thoughts. In July 2006, Dassault announced that Airbus had chosen the Enovia platform for all its aircraft development programs. "Airbus has decided to expand the usage of Enovia VPLM [virtual product life-cycle management] to all programs," Dassault said in a press statement. "The Catia and Enovia VPLM combination is becoming the standard environment for all new programs at Airbus."
Does that mean Parametric's Windchill must go? Airbus isn't saying, but management has a ways to go to straighten out confusion engendered by allowing different versions of the same CAD program, not to mention competing data management packages.
Meanwhile, Boeing's Dreamliner seems set to take off into much friendlier skies. Pittman and Fowler both say that perhaps the most important factor in ensuring the program's ultimate success was gaining executive support at the onset. Without this backing, they say it would have been impossible to ensure that partners and even all Boeing engineers were complying with demands to use the same software, update on schedule and save data when required.
"Mike Bair [787 program chief] and [Boeing CIO] Scott Griffin really understand the importance of PLM and stood behind Kevin [Fowler] and me," Pittman says. "That was crucial in ensuring compliance."
Dreamliner's "Virtual Takeoff"
In December, Boeing held a "virtual rollout" of the Dreamliner. It demonstrated how the aircraft has been designed and will be manufactured to about 3,000 employees and more than 100 visiting airline representatives.
As part of the demonstration, Boeing showed how in the early design stages, it was discovered that an electronics box manufactured by supplier Hamilton Sundstrand wouldn't fit into the plane's electrical equipment bay. The conflict was caught, and highlighted in red, by the Catia design software. Engineers were able to redesign the bay, essentially by shifting a beam, so the box would fit. If that conflict had not been caught until production began, it could have led to lengthy delays or a costly retrofit.
As impressive as the virtual rollout seemed, it was just that—a virtual rendition of the plane. A Dreamliner has yet to be built, and despite Fowler's confidence, Boeing cannot be certain it won't run into the same difficulties encountered by Airbus, or new, as-yet-undiscovered challenges involving the heavy use of composite materials in the plane, says Hans Weber, president of Tecop International, a San Diego aviation consulting firm.
Still, if Boeing succeeds in its drive to deliver its first 787 Dreamliner to All Nippon Airways in May 2008, seven months after the first A380 is set to go into service, it will have achieved a number of concrete benefits through the use of PLM software.
First, it will have shaved one year, or about 20% to 25%, off the time it normally takes to shepherd a new plane from concept through design and engineering and into production, according to Fowler.
As a prime example of how those savings are achieved, Fowler points to virtual test flights.
Just over a year ago, pilots began putting the 787 design through a variety of flight simulations. During one such test, where the plane is evaluated on whether it could take off on a single engine, pilots determined that it didn't have enough "fin control"—that is, fins used to stabilize the aircraft weren't functioning as desired. Over a period of four weeks, engineers evaluated more than 50 different fin configurations to give pilots the performance they were looking for. "Traditionally, we might have been only able to evaluate three or four options, and it would have taken three months," Fowler adds.
Second, Boeing expects to achieve a similar 20% to 25% cut in development costs by eliminating costly mock-ups and by working out kinks in manufacturing before going into production. Based on the plane's estimated $8 billion to $10 billion development budget, it could translate into savings in the area of $2 billion to $3 billion.
"One of the biggest lessons we learned on the 777 program," Fowler points out, "is that there is tremendous value in the digital data in terms of how it can be utilized throughout the entire life cycle of a project.
"That has been our biggest objective with the Dreamliner—to use that single source of data from when we take customer orders through design, through manufacturing and right through to support after the customer takes delivery," he says. "That's the real power and where the real benefits come from with PLM."
Editor's Note: This story was updated to explain how Boeing is using Parametric Technology's Windchill product lifecycle management software.
Boeing Base Case
Headquarters: 100 N. Riverside, Chicago, IL 60606
Phone: (312) 544-2000
Business: The world's largest combined manufacturer of commercial
airplanes and military aircraft.
Chairman, CEO: James McNerney
CIO: Scott Griffin
Financials in 2005: $54 billion in revenue, $2.6 billion in profits,
profit margin 5%.
Financials, first nine months of 2006: $44 billion in revenue, $1.2
billion in profits, profit margin 2.8%.
Challenges: Maintain lead over Airbus as world's largest commercial airline producer, and shepherd the revolutionary 787 Dreamliner into production.
- Maintain position as world's largest producer of commercial airplanes by increasing yearly airline orders by 284%, from 272 net orders in 2001 to 1,044 net orders in 2006.
- Shave 20% to 25%, or about one year, off the time it normally takes to develop a new plane, from six years in 1995 with the introduction of the 777 to five years with the planned launch of the Dreamliner in 2008.
- Reduce development cost of Dreamliner through the use of PLM software by 20%, from about $12 billion to $10 billion.
- Reduce time it takes to assemble a 787 in Everett, Wash., from about six days at start of production in 2008, to about three days after it builds the first 100 planes in 2010.