One speaker noted that one of his biggest prob-
lems is the datai— t is usually messy and often has
missing observations. We also heard about the oper-
Missile Reliability
ational point of view in a talk by a representative of
the Joint Forces Command. He gave us interesting
During engineering design, you expect to find failure modes.
information about the analytical challenges in the
For example, in a missile design, you might fire five missiles
joint operating environment, how the forces will be
and see two failures—say an engine ignition or battery failure.
fighting in the future, and how he sees statisticians
the engineers then try to fix the failures. After the fixes are
contributing to the nation’s security.
incorporated, how do you calculate the reliability?
In the question-and-answer period that fol-
lowed each presentation, participants asked ques-
to simplify, the original approach (using the Lloyd method)
tions and suggested ways statistics could help
turns failures into successes if those failures don’t occur again in
address the issues.
the next few shots. the method is simple, direct, and easily
Most speakers had a basic understanding of sta-
tistics, but came away with a better appreciation for
calculated. It led to a reliability estimate of about 0.93 in our
the role statistics can play. We tried to impress upon
case. the raw score would be the result if the failures were not
them the gap between the state of the art of statis-
turned into success, but retained as failures. Both methods are,
tics and what they may have picked up in a gradu-
in some sense, wrong. using the raw score fails to account for
ate statistics course 20–30 years ago. They learned
the effort (and possible success) to fix failure modes. Yet, the about areas where their interests overlapped those
Lloyd method gives too much credit to the fix, because it does
of statisticians. Contact information was exchanged
not recognize that observing one failure may prevent, or
and follow-ups are occurring.
shadow, the possibility of seeing another.
The afternoon session featured remarks by
SDNS members on how statistics was already being
what we have here is a system that has many failure modes,
used successfully on defense- and security-related
each with its own failure rate, and the failure rate of the whole
problems. These presentations included work on
system is a combination of all failure rates. when an engineer
biosurveillance, test and evaluation, and text data
removes a failure mode, the combined failure rate will
mining. Most presentations will be posted to the
change. Common experience is that each failure mode
SDNS web site at
www.amstat.org/sections/sdns.
Other activities members of SDNS have pur-
follows a Poisson distribution. so, the growth in reliability is
sued in an attempt to reach out to decisionmakers
nonhomogeneous Poisson process (more complicated
include hosting meetings between ASA presidents
because the events are discrete). the method is not simple,
and Department of Defense and CIA policy offices
not direct, and not easily calculated.
to raise the visibility of statistics. While the recep-
tion has been positive, it is important that this be an
In this case, the absence of a statistician in the beginning
ongoing activity and that we also seek meetings with
allowed both the government and contractor to be fooled
officials in the Department of Homeland Security
about the reliability of the system, illustrating our point
and Department of Energy Office of Science.
that statisticians have an important role in helping the We also have reached out to the Defense Science
government determine how best to use the available data.
Board (DSB), partly through ASA leadership, and
have been told that statisticians will be recom-
mended to DSB study panels that require statistical
expertise. We continue to push for a statistician to
Advanced Research Projects Agency (DARPA), and be appointed to DSB.
Joint Warfare Analysis Center.
Nozer Singpurwalla of DARPA gave the keynote
Outreach to Fellow Statisticians
address. He focused on the historical roots of statis- Recognizing the importance of having more stat-
tics in defense problems, applications of statistical isticians involved in national security projects to
approaches, and success stories. broaden our influence, we have undertaken many
Other presentations largely focused on the chal- activities to educate the community about the many
lenges the speaker’s office/organization encountered interesting problems and their potential for statis-
and where they thought statistics might be able to ticians. In addition to inviting statisticians to this
play a role. They addressed issues such as how good summer’s meeting, we’ve been organizing JSM ses-
they are at anticipating defense needs, whether the sions that show the breadth of activity in defense
systems they produce are better than what is already and securityf— rom cyber security to biosurveillance
available, and what is likely to be transforming. to more traditional areas such as reliability. We also
32 AmstAt News December 2009
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100