Are Boeing's Battery Difficulties Part of a Much Bigger Problem?
Seventy percent of the Dreamliner is preassembled by a tiered system of subcontractors that Boeing has never used before
Steven Cherry: Hi, this is Steven Cherry for IEEE Spectrum’s “Techwise Conversations.”
The Boeing Dreamliner has been in the news a lot, and not in a good way. But while Boeing’s reliance on the somewhat unproved technology of lithium-ion batteries has gotten a lot of attention, there were many other things about the Dreamliner that had never been tried before.
For example, the fuselage contained carbon-fiber-reinforced plastic, titanium, and only 15 percent aluminum. Each plane was supposed to accommodate engines from either Rolls-Royce or GE and be swappable between them in a mere 24 hours. The entertainment system was vastly expanded from previous planes, but also, for the first time, Boeing put it on the same network as the flight-control system. As the saying has it, “What could go wrong?”
Even more critically, Boeing’s other planes have a straightforward supply chain. The Dreamliner’s was different: Besides component suppliers, there were system suppliers, and then preintegration suppliers, called tier 1 strategic partners. So there were two additional levels in the chain that had never existed before. And whereas previous models were 30 to 50 percent outsourced, for the Dreamliner it was more than two-thirds. Again, what could go wrong?
A lot, say Christopher Tang and Joshua Zimmerman. They’re the authors of a rather prescient 2009 business journal article entitled “Managing New Product Development and Supply Chain Risks: The Boeing 787 Case.” [PDF]
We don’t know what will go wrong next, but if and when it does, it might be due to Boeing’s all-too-innovative supply chain. Here to explain it is one of those authors.
Christopher Tang has the title of distinguished professor in UCLA’s department of business administration. His areas of expertise include global supply-chain management, which he’s also taught at the Hong Kong University of Science and Technology, the National University of Singapore, and Stanford. He joins us by phone.
Chris, welcome to the podcast.
Christopher Tang: Thank you for having me.
Steven Cherry: The outsourcing model was unconventional, but it wasn’t entirely unprecedented, right? Toyota cars are produced using tier 1 strategic partnerships, and 70 percent of a Toyota car is outsourced. But you found some key differences between how Boeing and Toyota manage those relationships. First, maybe explain the multitier approach and then what made Boeing’s different from Toyota’s.
Christopher Tang: In terms of the multitier structure, [it] has existed before—that is how Toyota has been producing their cars for the last few decades. So Boeing has actually adopted that multitier structure, so by itself it is not a major problem. But I think that is because the product is brand new, that utilize a lot of unproven technologies, including the composite structure, including the entertainment system, including the lithium battery. So when you put all these unproven technologies together in a multilayered structure, that indicates that you could have more risks in the supply chain.
Steven Cherry: Now, I understand also that Toyota works extremely closely with its tier 1 suppliers, but Boeing kind of relied heavily on some technology called Exostar. Tell us about that.
Christopher Tang: Well, I think that when Boeing decided to change the multitier structure, they delegated the design of the products to the tier 1 suppliers, unlike the Toyota structure. Toyota cars are based on proven technology, and they have a well-established relationship with the tier 1 suppliers, such that they have a much closer relationship, much closer coordination and communication.
In Boeing’s case, when they decided to outsource the design, as well as the integration of the modules to the tier 1 suppliers, they really didn’t have a clear structure, communication strategy, coordination plan with the tier 1 structures, especially when they are located in different countries. So as a result, I think that Boeing decided to use the IT system Exostar to coordinate, to communicate with the tier 1 suppliers. But unless the tier 1 suppliers truly reveal all the truth to Boeing, otherwise the communication could be complicated.
Steven Cherry: So I guess it’s fair to say that Exostar was a sort of project-management software that was built specifically for the Dreamliner?
Christopher Tang: I don’t think so. It is a project-management software that has been used for other projects as well.
Steven Cherry: I see. And does Toyota use anything like it?
Christopher Tang: I think Toyota still relies on face-to-face meetings. Also, they might—the Toyota multitier structure is built on proven relationships in the past. If you study the Toyota supply-chain structure, you find out there’s most of the tier 1, tier 2, even tier 3 suppliers, they were either former Toyota trusted employees or former Toyota executives. So that’s why they call it a [unintelligible]. It’s like affiliated companies that they have long-term relationships, they have been closely communicated, closely coordinated, and closely trusted partners. Unlike Boeing’s case, when they delegate to the tier 1 strategic partners, they really don’t know who the tier 2 or the tier 3 suppliers were.
Steven Cherry: There was a US $1 billion decision to buy a key subcontractor. And that’s actually what led you to examine the Boeing supply chain back in 2009. It turned out to have been highly “unconventional,” to use the word you use in the paper. Tell us about Vought. It seems almost like a case study within a case study of what can go wrong.
Christopher Tang: Well, I think that in Boeing’s case that was an unintended consequences, when they decided to outsource to tier 1 suppliers, because they did not fully vet the suppliers’ capability. Then later on, they discovered that some of the suppliers are not capable to deliver what they had promised to deliver as a result. So Boeing has to take over the company so that they can control the development of the programs. So in this case, there were two companies they acquired subsequently. One of them is a tier 1 supplier called Vought—one of its divisions, they paid over $1.4 billion. The other one is called Global Aeronautica. They also paid over $1 billion for it. Because those two units, they were underperformed suppliers, so Boeing had to take over the control of the design so that they can really continue the development process.
Steven Cherry: And these were not isolated cases, right? In practice, Boeing had to send out hundreds of engineers all over the world to do the oversight that the middle tiers were supposed to perform on their own or that was supposed to be exercised by the tier 1 partner.
Christopher Tang: Well, that was, again, the unintended consequences, because after they discovered that the tier 1 had subcontracted to some of the tier 2 without notifying Boeing, and also some of the tier 2 suppliers in turn subcontracted to some of the tier 3 suppliers, then when there was a major delay, Boeing had to track and trace down who these tier 2, tier 3 suppliers were, and then they discover that some of them are not doing their job properly. So that’s why they had to send engineers, their own engineers, around the world, to track and trace and help these tier 2, tier 3 suppliers to [get] back on track so that they can deliver Boeing’s 787.
Steven Cherry: Did anyone at Boeing have any experience in managing a multitiered supply chain?
Christopher Tang: In this particular case, based on what I observed, I have doubts about their capability. You look at the original team that was leading their efforts to deal with the 787 program, most of the folks on the team, they were focusing on marketing and finance. So there were no really key members who truly had the expertise in supply-chain management. There are reasons behind it, because in the past, the supply-chain structure of Boeing is rather simple. They deal directly with all close to 1000 suppliers directly. So they have full control. Also, all their supplies are built to specifications. Boeing does the design for the 737 and 747, and then they have the design, and then they turn over the design to the suppliers so that the suppliers will produce the parts of the components according to specifications.
Steven Cherry: There was also risk sharing built into the contracts with the suppliers, but you argue in the paper that Boeing made some mistakes there as well.
Christopher Tang: Let me explain a little bit about the risk-sharing contract. In the old days, for the old planes, such as 737 and 747, Boeing would pay each supplier immediately after they finished their own task. But for the 787 program, Boeing was thinking they’d maybe want to reduce costs. So as a result, they want to share the risk with the suppliers in the following way: They issued a risk-sharing contract such that none of the suppliers would get paid until all the suppliers finished the task. So put in other words, the suppliers would not get paid for the development costs and other costs until they finished and assembled the first plane that was delivered to All Nippon Airways, which happened last year. So in that case, that was causing delayed payments for the suppliers.
Now this one, on the surface it may give the impression that Boeing is not paying its suppliers until everyone finished, but they did not anticipate the reaction of the suppliers. When the suppliers were thinking that they would not get paid until everyone finished the task, there’s no incentive for them to hurry up. So consequently, they may be delaying, or they may put more focus on helping Airbus projects so they can get paid right away. So in any case, they’re not really focusing on the 787 development program. This actually has been reflected in the latest events, that Vought was delaying, and actually Boeing ended up acquiring their entire units, because there’s no incentive for them to hurry up, because they won’t get paid.
Steven Cherry: Yeah, even back in 2009, when your paper was published, Boeing was already seeing some cancellations, because the plane was so far behind schedule. So how far behind schedule was the final first deliveries, and how much over the original budget did Boeing go?
Christopher Tang: For the delivery of the first 787 plane to All Nippon Airways, it was over three and a half years late. In terms of the development costs, it was over $10 billion over budget.
Steven Cherry: I gather that…I’m sorry, $10 billion over the budget?
Christopher Tang: Yes. It was over $10 billion over budget, because they had to acquire two tier 1 suppliers who were underperforming. So there’s one. Second, they had to pay some of the penalty to some of the suppliers. And third, they had to replace some of the planes that some of the airlines can no longer wait for the 787. They had to swap this with other planes.
Steven Cherry: You wrote in the paper that “on the surface it appears that Boeing’s fundamental problem was caused by its attempts to simultaneously take on too many drastic changes.” This isn’t rocket science—and by the way, Boeing has some rocket scientists. What’s the point of having decades of experience of making commercial aircraft if you do everything differently from everything you’ve ever done before?
Christopher Tang: Well, I think we have to take a step back. Why Boeing decided to launch a game-changer, they call it, the Dreamliner, the 787 program. I think at that time, in 2003, they were under the gun to do something after Airbus launched the A380, the jumbo. So at that time, Boeing had to figure out whether they use the traditional technology to give a newer look of the plane, or they wanted to develop an entirely different commercial aircraft that would give the passengers and the airlines a different experience.
Now, from the strategic standpoint, I thought that was brilliant, because the 787 aircraft can truly differentiate itself from all the other commercial aircraft, because one, it’s lightweight—the fuel efficiency. More importantly, that allowed all the passengers to fly any city pair around the world nonstop. Now this one is very important, because after September 11, most passengers, they are dragging their feet to take connecting flights, especially when you fly from one city pair, from, let’s say from the U.S. to Asia. Sometimes you have to stop over. So I think 787 has a long-haul flight such that passengers do not need to catch the connecting flight. It’s a value-add. Also, it’s a composite, it can increase the humidity, then the passengers will feel less tired. So in that case, that is really a great idea. So it’s not the plane that they promised they’d deliver. I think the problem is the execution plan—how they go about developing the plane.
Steven Cherry: So where are we at today? Do you think Boeing is fully back on track, or is there still some way to go?
Christopher Tang: In my opinion, I think Boeing has some more work to do. There’s two major issues I think they need to iron out right away. First, the lithium-ion battery, I think they are going to redesign, modify the design a little bit, to ensure that the fire would no longer be a threat. I think this one would push back for a few months at a minimum. The other issue they need to think carefully about is the composite material. This is the first time in commercial aviation history they use carbon fiber heavily on a commercial aircraft. The properties of all this carbon fiber up at 40 000 feet high is not very clear. Also, more importantly, the wings are developed by three different Japanese companies, and the wing is attached to the fuselage in the middle section of the plane using epoxy tape. Now, epoxy tape can sustain a certain turbulence, and also how it will work in the long haul, that still remains to be seen. So I think that one, they need to pay close attention for public safety issues.
In the long run, I think Boeing needs to pay close attention to the issue of inspection and maintenance at different airports for the 787 program. This is because they’re utilizing a lot of carbon fiber. They need to find easy ways to inspect for cracks on the aircraft. Unlike an aluminum plane, you can check for cracks. For carbon fiber, based on my understanding, they need either X-ray or ultrasound to check for cracks on the composite material. Then the question is, how can they utilize X-ray and ultrasound in different airports around the world? Can they actually do the maintenance and the inspection effectively?
Steven Cherry: Yeah, the wing looked especially interesting to me in a diagram in your paper, that basically three different tier 1 suppliers are involved in that part of the plane, and it does seem a little untried.
Well, Christopher, it’s a fascinating backstory to a story we thought we all knew. Thanks for your research and for joining us today.
Christopher Tang: Thank you very much.
Steven Cherry: We’ve been speaking with UCLA distinguished management professor Christopher Tang about the problematic supply chain behind some of the problems with Boeing’s Dreamliner 787.
For IEEE Spectrum’s “Techwise Conversations,” I’m Steven Cherry.
NOTE: Transcripts are created for the convenience of our readers and listeners and may not perfectly match their associated interviews and narratives. The authoritative record of IEEE Spectrum’s audio programming is the audio version.