Engineering Professionalism

ENG101

Case Studies for the Design Process

Automotive Components, Safety
Testing case
Should he reveal everything (his belief
that the testing was inadequate and his
recent discovery) and cost the company a
great deal of money? Or should he testify
that he had been convinced that the
testing was adequate? Should he keep it
to himself that he now knew that there
was something wrong with the design?

2
Possibilities

• Charlie should tell everything but as a

middle ground approach. So, he should
stress that at the time, his team thought
the testing was adequate but he should
also explain his process of finding the
new problem and how to fix it so the
problem can be resolve. This way,
hopefully, Charlie can stabilize the
damage to himself and his company and
the people as much as possible.

3
• Charlie should testify that he did not feel
comfortable with the original testing as
well as sharing his new discovery. His 1st
obligation is to the safety of the public
and if his company is responsible for
someone’s death, it is his duty to be
honest and avoid keeping secrets.

4
• Charlie should tell that the initial testing
was believed to be adequate and that
once he heard of the issue he
reexamined the design and figured out
how to fix the issue. If he says that he
told his supervisors more testing was
needed he would likely loose his position
with the company.

5
• He should tell his company about the
solution to the problem. He should state
under oath that they were confident that
the sensor worked as it left the factory,
but he believed that further testing
would probably have been a good idea.

6
• Charlie should be honest and explain that
he had doubts about the testing. He
should not feel required to release the
design flaw he later discovered. If that is
brought up in court, he could argue that
it would have been revealed with more
testing.

7
Case: Honesty in
Specifications/Bidding

Arthur is chief engineer in a components house. As

such, he sits in meetings concerning bidding on
contracts. At one such meeting between top company
executives, who are interested in getting a major
contract, and the National Aeronautics and Space
Administration, NASA presents specifications for
components that are to be several orders of
magnitude more reliable than the current state of the
art. The components are not part of a life-support
system, yet are critical for the success of several
planned experiments. Arthur does not believe such
reliability can be achieved by his company or any
other, and he knows the executives feel the same.
8
Nevertheless, the executives indicate an interest to
bid on the contract without questioning the
specifications. Arthur discusses the issue privately
with the executives and recommends that they review
the seemingly technical impossibility with NASA and
try to amend the contract. The executives say that
they intend, if they win the contract, to argue mid-
stream for a change. They remind Arthur that if they
don't win the contract, several engineers in Arthur's
division will have to be laid off. Arthur is well liked by
his employees and fear the lay-offs would affect some
close friendships. What should Arthur do?
9
Challenger Disaster

• Challenger was a delta-wing craft with a huge

payload bay
• Three main engines fuelled by several million
pounds of liquid hydrogen
• Fuel carried by an immense external divided
fuel tank
• During lift-off main engines fire for 8.5 min.,
thrust for first two minutes provided by two
booster rockets

10
• Booster rockets fuelled by solid fuel
• Casing of each booster rocket about 150 ft long
and 12 ft in diameter
• NASA+companies: Rockwell Int. (orbiter and
main rocket), Morton-Thiokol (booster rockets)
• After embarrassing delays, flight set for Tues.
morning Jan 28, 1986

11
• A.J. McDonald of Morton-Thiokol worried about
predicted freezing temperatures since he knew
of problems with field joints on a previous cold-
weather launch
• He arranged a telecon
• A. Thompson and R. Boisjoly, seal experts at
Morton-Thiokol, explained how booster rocket
walls bulge and combustion gases can blow past
one or both of the O-rings that make up the
field joints

12
• O-rings char and erode as seen on many
previous flights
• In cold weather problem is aggravated
because rings and putty packing a less
pliable
• Past flights showed that below 65 deg
launches always resulted in failure
incidents with O-rings

13
• Engineering managers agreed there was
a problem with safety
• During a recess in telecon, Senior VP
Jerry Mason turned to engineering
managers and told him “to take off your
engineering hat and put on your
management hat.”
• Tone: You have to prove to us that there
will be a problem if we launch

14
• Countdown ended at 11:38am
• Temperature was 36 deg.
• Smoke came from field joint on take-off
• Soon turned into a flame
• Hydrogen in tank caught fire, tank broke
loose and smashed into Challenger’s
wing
• By 76 sec. into flight at 50,000 ft totally
engulfed in fire ball
15
• Died: F. Scobee, M. Smith, G. Jarvis,
R. McNair, E. Onizuka, J. Resnick, and
Christa MacAuliffe (“teacher in a
space”)
• What could have been done
differently?
• Clearer presentation of the problem
(temp influence on O-ring failure)? Could
have made for more forcible arguments?
• Watch for growing problems!
• Don’t take off your engineer’s hat!
16
Aerospace engineer: “The arrogance that
prompts higher-level decision makers to
pretend that factors other than
engineering judgement should influence
flight safety decisions and, more
important, the arrogance that rationalizes
overruling the engineering judgement of
engineers close to the problem by those
whose expertise is naïve and superficial by
comparison”

17
The Flaw in the Intel Pentium Chip
(case study by C. Fleddermann)

• In late 1994 media started reporting flaw in

Intel’s Pentium microprocessor
• It was the chip used in 80% of all personal
computers in the world
• Flaws in the integrated circuits of
microprocessors are not uncommon (most
of these cannot be detected by the user and
do not affect operation)
• The 1994 flaw was different. It caused
incorrect answers when performing double-
precision arithmetic (common operation,
easily detectable)
18
• Intel response:
• Acknowledged error but said that defect was
insignificant and the vast majority of users
would never even notice it
• Chip would be replaced for free only for
users who could demonstrate that they
needed an unflawed version
• Users found this unsatisfactory
• IBM, a major pentium user, cancelled the
sales of all computers using the chip

19
• After much negative press, and an outcry
from Pentium users, Intel agreed to replace
the faulty microprocessor with an unflawed
version for any customer who asked to have
it replaced
• Note: Long before news of the flaw
surfaced, Intel was aware of the problem
and corrected it on subsequent versions
• But, they continued to sell the flawed chip
• New Intel policy: Flawed chips should be
replaced on request, regardless of how
insignificant the flaw is
20
• Public relations problem, with ethical
issues
• Questions (Fleddermann):
• Should flaws always be revealed to
customers?
• Is it an ethics problem only if safety is
involved?
• What if they added a label “This product may
contain unexpected flaws and might not
operate correctly under all conditions”. Does
this solve the ethical problems for the
company?

21
• How can an engineer be sure that there are
no defects in a product? Testing! Before/after
product release
• If it is impossible to eliminate all defects in a
product, what level of defects is acceptable?
• Does this depend on the type of product?

22