Why no new Jumbo anytime soon

From:         mweber@t140.aone.net.au (James Matthew Weber)
Date:         21 Oct 1998 16:51:01 GMT
Organization: Kapor Enterprises, Inc.
View raw article
  or MIME structure

About 18 months ago I posted an article very similar to this. With the
current discussion about the future of the A3XX, I think the issues
raised are significant, and of interest, so I have essentially
recreated that article.

As the original article pointed out, the early estimates of costs by
Airbus were grossly unrealistic. I think the original cost was in the
3 billion USD range, most people in the Industry believe the real cost
was likely to be closer to 10 billion USD, and recent posts in this
group suggest this is still an accurate assessment. If you look at the
history of airliners, at this point, you market them for 5-6 years,
and then have to offer a major upgrade. The 747 has undergone several
such upgrades, Airbus aircraft have a similar history. While these
upgrades are not as expensive as the original development costs, they
are not cheap either.

 Realistically, if the program is going to be profitable, you need to
recover the R&D costs in the first 5-6 years of delivery. Given that
by Treaty, some 70% of the development costs must be met from
Commercial sources, that means the real R&D cost with interest will be
close to 13 billion USD. IF you assume you can sell 500 aircraft (and
it is not at all clear that the market in that period is that large.
(Boeing at its  best  only delivered about 60 747's per year) during
that period, that puts an R&D cost per aircraft at about 25 million
USD. If you sell fewer aircraft, the problem gets worse.

For an aircraft that needs to sell for probably no more than 250
million USD, that is a big piece of the costs. For the past decade the
capital cost per seat on aircraft have not gone down very much as the
size of the aircraft increases. A 747-400 is not a lot less than 3
times the cost of a 737-300!

Airlines want an aircraft that has lower direct operating costs. At
this point, that is a difficult problem. There are only a few places
you can work to reduce those costs. You can reduce the fuel burn. This
is entirely within the realm of possibility. The GE90 is substantially
more fuel efficient than the Trent or PW4000. (at full thrust on the
ground, it is about 6%, probably less than that at cruise). This was
done however with a sizeable weight penalty. A GE90 weighs some 3000
pounds (about 20%) more than a Trent! The fuel burn is such that if
you make the stage long enough, the fuel efficiency more than covers
the weight. Unfortunately it takes a very long stage to do it, on a
777 it happens at about 10 hours. Up that point however, the higher
engine weight reduces payload, and revenue, driving costs up rather
than down!

The market for aircraft that are attractive to operate only on stages
that are longer than 10 hours is probably rather limited, and even at
that, the gain in direct costs is far less than the 10-15% that
airlines are looking for. The vast majority of flights are a lot less
than 10 hours duration!

The other place to pick up economies is in the airframe. Unfortunately
this is much easier said than done. Since the late 1970's, sufficient
computer power and hydrodynamic codes have been available to allow
existing airframes designed since then to  be pretty much optimized
within the certification limits. You can see that  with the engines.
Until quite recently, everyone's best engine had a specific fuel
consumption of about .31. 
(CFM56, CF6, PW4000, RB211, Trent).  The GE90 does much better, but
the weight penalty is sufficiently high that it is attractive only on
the longest routes.

You can reduce the weight of the airframe, however composites, and
other exotic materials  have tended to substantially increase the
costs of the aircraft. These materials are more expensive to buy, and
more expensive to manufacture with than aluminum, and can be more
expensive to own.

  As a senior manager for Singapore Airlines put it, a single minor
error in servicing the fin on an A 320 costs upwards of $100,000 to
fix, and wipes out the entire fuel savings from the composite fin over
the entire life of the airframe. It will be expensive to make, and
expensive to fix if you damage it. That is why composites and exotic
alloys like Al-Li have seen such limited use to date. Obviously
increased costs associated with such exotic material will eat into any
direct operating costs savings they produce by increases the capital
costs. Capital costs are a substantial portion of total operating
costs. A thumbnail calculation put them at about .01USD per available
seat mile on a 747-400 (for anyone interested, I can provide the
details of the calculation). That is about 15% of typical cost per
available seat mile, so increasing the capital cost is painful to

Labor is a significant cost, however building airplanes has always
been a labor intensive process. Processes that decrease the labor
content tend to increase the capital costs however. Large gains from
this area are unlikely, and given European labor laws, have tended to
be problematic for Airbus. It's hard to reduce your work force.

The real gains will come from changes in the certification
requirements. Today we require that the airframe be unconditionally
stable. We do that by making sure we have plenty of drag in the right
places (a big tail and fin). It should be obvious to the most casual
observer that drag and efficiency are at cross purposes.  I don't see
a change in the certification requirements anytime in the next 10
years however. If you can get the certification requirement changed so
that unconditional stability is not required, all bets are off. I
haven't seen estimates of the potential gains on the civilian side,
however a comparison between the B1 and the B2 bombers is perhaps
instructive. The B2 has about the same payload as the B1, longer
range, and weighs about 30% less than a B1. Obviously there is
considerable gold to be mined if you can get the rules changed. That
isn't going to happen overnight, and no airline will buy into an
aircraft that cannot be certified within the rules that are on the
books today, so A3XX has to be built within current rules.

The nightmare is you build the aircraft, in 10 years, the rules do
change. That might  be only  5-6 years into the life of the A3XX,
leaving Airbus in much the same situation Lockheed ended up within
the Starliner program in the late 1950's ( a turbo prop
Constellation), Good airplane, but the 707 rendered it obsolete before
they could deliver very many of them, and those that were delivered
didn't stay in service for long!

My own suspicion is that Boeing is banking on this happening, and as a
result doesn't see a good reason to take the risks. Airbus wants the
market, and maybe prepared to take the risks if it isn't their money
at risk.  However if it really is the Airbus partners at risk for
about 10 billion USD, I am sure they will look long and hard at the
risks. Once the rules change, it will be possible to build a far more
efficient air frame. No one wants to be in a position where they have
a large investment in obsolete technology when it happens however.

At this point all airlines are interested in A3XX, but until or unless
Airbus can clearly demonstrate that it can in fact produce the 10-15%
reduction in direct operating costs with capital costs that are
competitive with the 747-400, I don't think anyone will be signing on
the dotted line. At the end of the day, the inability to produce the
product airlines wanted at a cost that was acceptable to Boeing and
the customers is what put the 747-500/600 back on the shelf. The costs
for those programs were a whole lot less than the likely costs of the
A3xx program. Boeing clearly has recovered the capital costs on the
747 program, and could probably afford to sell the 747 for a good deal
less than the current price. Doing so would obviously put pressure on
Boeing's profits, but might well  be a near  fatal for the A3XX

The problems are real, and many of the risks are beyond Airbus's
direct control, and one is clearly in the hands of the competition.
The risks are not t trivial, and that is why none of these programs
are going anywhere very fast!

I'd add one other seeming unrelated comment: The capacity constraints
at Japanese airports are the result of political decisions rather than
technical decisions. Attempting to apply technical solutions to
political problems traditionally doesn't work very well. The solution
to the capacity problems at places like Narita is to solve the
political problem, and add the runway capacity. It has to be less
expensive than designing new airplanes! Much of the service to and
from these airports isn't long haul. The 747-SR and 747-400D are used
only because there isn't anything else available, not because they are
well suited for the job!  The A300/330-300 are  well suited for that
type of service, and are very popular in Asia. Boeing has never really
built direct competitors for either aircraft.

Most of the things you do to aircraft to give it good short haul
performance are not good for long haul operation and vice versa. 

My thoughts and  opinions anyway.