快猫短视频

THE International Space Station will be completed up to three years late and
cost American taxpayers $7.3 billion more than its current $17.4
billion price tag, according to an official audit of the NASA-led project. Yet
even this grim assessment seems unduly optimistic, 快猫短视频鈥檚
calculations reveal.

Last September, under mounting pressure from Congress, NASA administrator
Daniel Goldin called for an independent cost assessment of the station from a
task force led by Jay Chabrow, a specialist in aerospace cost analysis and
president of JMR Associates, a technology consulting firm in Las Vegas.
Chabrow鈥檚 report, released late last month, suggests that NASA has got its sums
badly wrong.

But even this analysis doesn鈥檛 account for problems caused by launch vehicle
failure, beyond noting that 鈥渢here is a high likelihood that one or more
failures, including catastrophic failures, will occur鈥�.

快猫短视频 has calculated the probability of various kinds of
failures (see 鈥淗ow the numbers stack up鈥�). The probability of at least one
launch failure resulting in the loss of its cargo鈥攖hrough an explosion,
placement in the wrong orbit or another mishap鈥攊s 99.5 per cent. In fact,
NASA and the Russian Space Agency should together expect to lose five missions,
with the Russian rockets being most vulnerable.

Most of the 60 Russian launches will carry general supplies, so their loss
would be inconvenient, but not catastrophic. But the probability of losing one
of the 33 shuttle missions or 12 Russian launches carrying essential hardware
for the station is still an alarming 73.6 per cent.

Such an event could add years to the station鈥檚 assembly timetable and several
billions of dollars to its cost. 鈥淚t really depends on what element fails and
where in the sequence you are,鈥� says Michael Hawes, chief engineer for the
station at NASA鈥檚 headquarters in Washington DC. 鈥淵ou might have a full delay of
the manufacturing time鈥攎aybe two to three years.鈥�

Hawes says that some parts of the assembly sequence depend on the previous
step being completed, so a delay in one segment could hold up the whole project.
This would be the case if one of the truss segments that will make up the station鈥檚 backbone
(see Diagram)
were to be lost early in the assembly sequence.

There are backups for some of the elements. The US Navy, for instance, has a
module that could be pressed into action instead of the Russian-built service
module, which was due to fly on a Soyuz rocket in December, but is now
delayed.

If the Russian module were lost, it could be many months before its stand-in
hits orbit. And as it is currently engineered, the Navy module can house
astronauts for only short periods鈥攔ather than the permanent occupation
envisaged for the service module. This would introduce further delays to the
assembly schedule.

As the station nears completion, NASA and its international partners would
probably cut their losses if faced with a failure. 鈥淎fter a certain point, once
you鈥檝e got the things you couldn鈥檛 be without, you鈥檝e got a functioning space
station,鈥� says Michael Peters of the US Air Force Cost Analysis Agency in
Arlington, Virginia, a member of Chabrow鈥檚 task force. 鈥淚f there鈥檚 a failure,
maybe we鈥檒l say: that鈥檚 it.鈥�

Asked to comment on 快猫短视频鈥檚 analysis, NASA officials
questioned the launch reliability estimates. But even assuming NASA鈥檚 estimate
of 99.6 per cent for the shuttle鈥檚 reliability since the Challenger disaster,
and a generous 93.2 per cent for the Russian rockets, taken from published
sources, the chance of losing one assembly mission is 62.3 per cent. With the
supply flights, it becomes 98.7 per cent.

鈥淚t鈥檚 not an unusual or abnormal problem when you鈥檙e launching a huge
constellation of satellites,鈥� says Peters. But unlike NASA, commercial
organisations planning extensive launch sequences have accounted for rocket or
other hardware failures from the beginning. Motorola leads the consortium of
companies behind the planned Iridium constellation of 72 communications
satellites, involving 15 launches. 鈥淲e factored failures into our planning,鈥�
says Kathi Haas, a spokeswoman for the company鈥檚 satellite systems division in
Chandler, Arizona.

Chabrow wants NASA to do the same: 鈥淪omebody should take that into
肠辞苍蝉颈诲别谤补迟颈辞苍.鈥�

EVERY time we take a flight, drive a car or cross the street, we鈥檙e taking a
risk. Luckily, most of the things we do regularly are pretty safe. But if you
take a moderate risk, like launching a rocket, many times, the danger increases
very quickly.

According to the independent space station auditing team led by technology
consultant Jay Chabrow, good rocket launchers deliver their payloads
successfully around 92 per cent of the time. 鈥淲e鈥檙e not saying there鈥檚 anything
wrong with the launchers,鈥� says Chabrow.

Launch a Russian Proton or Soyuz rocket 12 times, and the probability that
there will be no major hitch is 0.92¹² which works out as 0.368,
or 36.8 per cent.

The space shuttle is more reliable, performing so far at just under 99 per
cent. Include the 33 shuttle launches needed to complete the space station鈥檚
assembly and the likelihood of all the pieces surviving is
0.92¹² 脳 0.99³³, or
26.4 per cent. Add in 48 more Russian supply missions, and the probability of
reaching the correct orbit is just
0.92⁶⁰ 脳 0.99³³, or 0.5 per cent. That鈥檚
less likely than flipping a coin seven times and getting seven heads.