Boeing Delays 787 First Flight

 - September 6, 2007, 7:16 AM

Boeing now estimates that first flight of the 787 will happen sometime between mid-November and mid-December–at least three months later than originally planned–due to delays in coding flight-control software and completion of so-called “traveled work”–tasks originally meant for partners but passed on to Boeing’s final assembly facility in Everett, Wash. Speaking yesterday during the company’s latest formal briefing on the program, Boeing executive vice president and Commercial Airplanes CEO Scott Carson said that the revised schedules still supports plans to deliver the first production airplane to All Nippon Airways in May, but that the delay means “added risk for sure.”

Program manager Mike Bair elaborated on the nature of the risk in fairly unambiguous terms. “The real issue is if we have some discovery in the flight-test program that causes us to have to go back and do some sort of redesign or rework of the airplane,” said Bair. “We’re rapidly running out of time to be able to deal with anything big.”

In effect, Boeing has more or less exhausted what Bair called the time buffer it had built into its schedule to deal with surprises during flight testing. “It’s really tight,” Bair conceded. “We’ve got some other ways to add buffer by eliminating testing in the flight-test program. That’s not something we want to do, but it’s getting tighter and tighter as this thing pushes into the fall.”

Although the backlog of traveled work on the first airplane stemmed mainly from a global shortage of fasteners, Boeing’s subsequent discovery that documentation accompanying the structures sent to Everett did not correspond with the work left to do on the airplane compounded the problem. That failure required Boeing first to spend time straightening out the documentation and, second, to sequence the remaining workflow properly.

Boeing recently told several of its structural partners to rearrange their ship dates starting with the second flight-test airplane and to finish all systems, wiring and other installations before shipping any structures to Everett. The change means that the third airplane scheduled for final assembly, just behind the static test airplane, will now serve as the fatigue test airplane. The fourth airplane on the line, scheduled for assembly in October, will serve as flight-test airplane number 2. 

Meanwhile, both Boeing and Honeywell, which supplies the flight control system software, have added manpower to ensure that the program’s system integration labs receive properly coded software for the flight controls by the end of this month. “We pushed really hard to complete coding of our flight control system software in August,” said Bair. Although he said Honeywell carries responsibility for “the bulk of the work,” Bair avoided assigning any blame given the rather large role Boeing plays in integrating the software.

In any case, Boeing cannot fly the first prototype until it replaces all the temporary fasteners used to ship structures to Everett and present an assembled airframe for rollout on July 8. At last count, said Bair, the airplane still contained some 700 non-airworthy fasteners–a marked improvement over the multiple thousands present for the rollout. “We are being very methodical with [fastener supplier Alcoa] on a daily, hourly basis on prioritizing what fasteners need to be produced and when they need to be produced,” said Bair. “This fastener shortage heals basically one lot of bolts at a time.”