The Unintended Consequences
of a career in Engineering
Or
How to end up a mass murderer
without even trying
What am I here to talk about?

The PATRIOT System, and my part in it





Stands for Phased Array Tracking, Radio
Intercept Of Target
Designed and built at Raytheon, right here in
Massachusetts
First built as an anti-aircraft missile system
Modified to function as an Anti-Ballistic Missile
system
Deployed during the First Gulf War in Saudi
Arabia, Kuwait, and Israel
Here's what it looks like
How PATRIOT was used


Initial interceptions of Scud missiles were very
successful, but they used 4 PATRIOT missiles per
Scud – just to make sure
This was a problem:




Iraq had hundreds of Scud missiles, more than all
the PATRIOT missiles ready to use.
So, to make interceptions more economical,
modifications were made to have the PATRIOT
disable the Scuds, rather than destroy them
Every effort was made to find and destroy the mobile
launchers, without which the Scuds were useless
PATRIOT batteries were permanently placed
around sensitive areas, manned and running 24/7,
in case of missile attack
How effective was it?


In the initial trials, extremely
In 'economy' mode, somewhat less effective






The strikes by the PATRIOT missiles were skewed
towards the front of the Scud
The idea was to disable / detonate the warhead
Worked as planned, mostly, but that left 2 tons of
rocket still falling on the target
It helped that the Scuds could not be aimed well
When deployed to Israel, the 'economy' mode was
less helpful as Israel is a densely populated area.
The worst failure occurred Feb 25, 1991
So what happened?
February 26, 1991
WAR IN THE GULF: Scud Attack; Scud Missile Hits a U.S. Barracks, Killing 27
By R. W. APPLE Jr., Special to The New York Times
In the most devastating Iraqi stroke of the Persian Gulf war, an Iraqi missile demolished a barracks housing more than 100
American troops on Monday night, killing 27 and wounding 98, the American military command in Riyadh said early today.
Some of those who lived there were women, but a military spokesman said he did not know whether any women were killed or
wounded. A pool report said the barracks housed the 475th Quartermaster Group, an Army Reserve unit based in Farrell, Pa., a
small town near the Ohio state line.
A television correspondent said he had seen a Patriot defensive missile intercept the Soviet-built Scud, a missile that has been
fired again and again into Saudi Arabia or Israel during the war, often on erratic courses. Monday night's attack was the first of the
Iraqi surface-to-surface missiles to take a sizable number of lives.
Greg Siegle, a freelance writer, said he saw "a gigantic explosion" 100 feet off the ground, then another as bits of glowing metal fell
to the ground and set fire to the barracks, which is situated in Al Khobar, a city a few miles from Dhahran and its big air base. He
said the building caught fire at once. Within an hour, it was a charred skeleton.
The American command disputed the accounts of the witnesses. It did not say whether a Patriot missile had been fired, but said
the Scud had broken up in flight, as have several others launched in the last few weeks.
Baghdad radio, hailing the attack, said the missile struck "the coward traitors who mortgage the sacred places of the nation and
turn Arab youth into shields of flesh."
Many of the occupants of the barracks, a corrugated-metal warehouse that had been converted into temporary housing for British
and American troops, were eating dinner or relaxing when the missile hit. Others were apparently working out or sleeping, and
many survivors wandered around, in sweatsuits or gym shorts, stunned, during the rescue efforts.
A military policeman said that most of those in the building were Americans from Army transportation and quartermaster units. Not
all of the names of the units were immediately available.
In Pennsylvania, an operator taking calls at the Farrell Police Department last night said that members of the 475th Quartermaster
Group came from Farrell and many other towns along the Ohio-Pennsylvania border. The operator said she had seen several
families gathered outside the group's headquarters at the United States Army Reserve Center on her way to work.
Chaos engulfed the scene moments after the burning debris fell into the converted warehouse. Saudis in the neighborhood
followed it down from the sky and quickly ran to the scene, complicating rescue efforts. Mr. Siegle said it took at least 20 minutes
for the first fire engines to arrive.
GAO
United States
General Accounting Office
Washington, D.C. 20548
GAO's Account
Information Management and Technology Division
B-247094
February 4, 1992
The Honorable Howard Wolpe
Chairman, Subcommittee on Investigations and Oversight
Committee on Science, Space, and Technology
House of Representatives
Dear Mr. Chairman:
On February 25, 1991, a Patriot missile defense system operating at Dhahran, Saudi Arabia, during Operation Desert Storm failed to
track and intercept an incoming Scud. This Scud subsequently hit an Army barracks, killing 28 Americans. This report responds to your
request that we review the facts associated with this incident and determine if a computer software problem was involved. If so, you
asked that we provide information on what the specific software problem was, and what has been done to correct it. Appendix I details our
objectives, scope, and methodology.
Results in Brief
The Patriot battery at Dhahran failed to track and intercept the Scud missile because of a software problem in the system's weapons
control computer. This problem led to an inaccurate tracking calculation that became worse the longer the system operated. At the time of
the incident, the battery had been operating continuously for over 100 hours. By then, the inaccuracy was serious enough to cause the
system to look in the wrong place for the incoming Scud.
The Patriot had never before been used to defend against Scud missiles nor was it expected to operate continuously for long periods of
time. Two weeks before the incident, Army officials received Israeli data indicating some loss in accuracy after the system had been
running for 8 consecutive hours. Consequently, Army officials modified the software to improve the system's accuracy. However, the
modified software did not reach Dhahran until February 26, 1991--the day after the Scud incident.
So what went wrong that day?



A software error in the Engagement Control
System caused the PATRIOT guarding the
area to lose track
Without a good target track, the PATRIOT
missiles cannot engage
So it never fired at the Scud
Normal operation
The range gate error
What happened that day
A Perfect Storm of
circumstances


The PATRIOT system's software had a flaw
Our Army never found the flaw due to how
they used the system




Used as a point defense system against the
Soviet Union mostly
Moved around constantly for security reasons
Pretty reliable, but if down, it took a long time to
repair
I had just finished the TMU project at
Raytheon, and the upgraded units were in
the field.
The Storm continues




The PATRIOT systems were being used in
fixed positions, run 24 hours a day
With the TMU improvements I made, the
system was both more reliable (it could spot
trouble sooner) and much quicker to repair
The Israel Army found the flaw in PATRIOT
when running for as little as 8 hours
They alerted our Army, which issued a vague
warning that didn't prompt anyone to action
The Storm breaks






The Iraqi launched a Scud missile at
Dhahran, in Saudi Arabia
The Scud functions perfectly
The PATRIOT battery guarding the city was
running too long, and the flaw prevented the
missile from being tracked.
The automatic systems never fired
There was no time for the operators to do
anything
The Scud hit the barracks, killing and
wounding 125 people
And yet it could have been
prevented




The Army worked on a fix to the flaw for 2
weeks after it was found
Distributed patches for the software to all
PATRIOT units in the theater
The patches arrived on Feb 26th
One day after the barracks was destroyed.
Nothing like that can happen to
me, right?
•
•
•
No matter what we're working on, there is
almost always some potential for disaster,
especially with software and hardware for
the military (also CIA, NSA, etc.),
transportation, and medicine (and many
others).
Sometimes it's not even doing your job
wrong, but doing it right.
How will YOU handle it?