3 ms·

You're missing the point: of course it's not in BP's interests to have a well explode, but the people making the important decisions are often not making those

You're missing the point: of course it's not in BP's interests to have a well explode, but the people making the important decisions are often not making those decisions based on sound engineering, but on political pressures and incentives from above. Richard Feynman showed an excellent example of this after the Challenger disaster. He was on the Presidential task force after that disaster, and he asked NASA management and engineers, separately, to quantify what they felt the odds of a catastrophic failure resulting in crew loss were prior to the disaster. The risk values quoted by engineers and by management were something like 2-3 orders of magnitude different (with the engineers quoting the much more likely value)...

gvb 16y ago
"Engineers at Rocketdyne, the manufacturer, estimate the total probability [of catastrophic failure] as 1/10,000. Engineers at Marshal estimate it as 1/300, while NASA management, to whom these engineers report, claims it is 1/100,000. An independent engineer consulting for NASA thought 1 or 2 per 100 a reasonable estimate." — Personal observations on the reliability of the Shuttle by R.P. Feynman http://science.ksc.nasa.gov/shuttle/missions/51-l/docs/rogers-commission/Appendix-F.txt http://science.ksc.nasa.gov/shuttle/missions/51-l/docs/roger... Empirical post script (as of STS-132 2010-05-26) Total launches 131 Failures 2 Successes 129 Failure rate: 1.5% If you read the report, you will find that NASA declared a reliability number that the shuttle was to meet (1/100,000), and then backed into the calculations that justified that reliability number. In other words, it was top-down, rather than bottom-up. IMHO, bottom-up tends to be pessimistic, but top-down tends to be be a figment of management's imagination. In his report, Dr. Feynman pointed out that many of the calculations had reliability figures that, while not necessarily wrong, were so small as to be impossible to validate... but that was the number necessary to generate the 1/100,000 roll-up (actually, rolled-down) number.

gvb 16y ago

"Engineers at Rocketdyne, the manufacturer, estimate the total probability [of catastrophic failure] as 1/10,000. Engineers at Marshal estimate it as 1/300, while NASA management, to whom these engineers report, claims it is 1/100,000. An independent engineer consulting for NASA thought 1 or 2 per 100 a reasonable estimate." — Personal observations on the reliability of the Shuttle by R.P. Feynman http://science.ksc.nasa.gov/shuttle/missions/51-l/docs/rogers-commission/Appendix-F.txt http://science.ksc.nasa.gov/shuttle/missions/51-l/docs/roger... Empirical post script (as of STS-132 2010-05-26) Total launches 131 Failures 2 Successes 129 Failure rate: 1.5% If you read the report, you will find that NASA declared a reliability number that the shuttle was to meet (1/100,000), and then backed into the calculations that justified that reliability number. In other words, it was top-down, rather than bottom-up. IMHO, bottom-up tends to be pessimistic, but top-down tends to be be a figment of management's imagination. In his report, Dr. Feynman pointed out that many of the calculations had reliability figures that, while not necessarily wrong, were so small as to be impossible to validate... but that was the number necessary to generate the 1/100,000 roll-up (actually, rolled-down) number.