Newsstand - More hand searches for airline bags

More hand searches for airline bags

Bomb-detection machines are too prone to false alarms

The automated bomb-detection machines being used to screen luggage at U.S. airports are so prone to false alarms that federal officials plan to privately search many checked bags by hand.

http://www.msnbc.com/news/830759.asp

“We want to make sure we have a process in place that would allow for travelers to be reassured of the safety of their items,” said Leonardo Alcivar, a spokesman for the Transportation Department, which oversees the TSA. “It’s another sea change in how security is conducted.”

Maybe you idiots could spend the money that you're currently squandering on "random security" on developing a bomb detection that actually works? Nah... too simple I guess.

------------------
"Give me Liberty or give me Death." - Patrick Henry

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Spiff:
Maybe you idiots could spend the money that you're currently squandering on "random security" on developing a bomb detection that actually works? Nah... too simple I guess.</font>

Maybe you'd like to know how much money has already been spent to develop the existing equipment before criticizing.

Then again, your use of the word "idiot" already tells us plenty about what you think.

Does it really matter how much money has been spent? The equipment does not work -- that is well established. The error rate (false positives) is somewhere around 30 percent. What more do you need to know?

If I spent a billion dollars trying to turn lead into gold (like the ancient alchemists), should I be deemed to have succeeded simply because I spent a lot of money? Come on, you can do better than that!

Bruce

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Plato90s:
Maybe you'd like to know how much money has already been spent to develop the existing equipment before criticizing.</font>

No, I wouldn't. I will probably sleep better if I don't know how much has been wasted, eh, spent, on it. And it really doesn't matter if they spent $5 or $5,000,000,000. As long as it doesn't work, it doesn't work. But we still have to use them. Aren't political decisions nice? Especially in an election year.

The infuriating part of this is that Congress was told last September-November that the stupid devices were prone to about a 25% false positive rate yet the Morons fired anyway, hoping later that the aim was ok. http://www.flyertalk.com/forum/mad.gif

Vote today to send all the morons in Congress home. http://www.flyertalk.com/forum/mad.gif

Don't bomb sniffing dogs have a better rate of accuracy? And for the $millions spent on each machine, the airports could alreay have a whole pack of cute tail wagging dogs on duty. And it would take a whole lot shorter time frame to train and deploy the dogs.

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by onedog:
Don't bomb sniffing dogs have a better rate of accuracy? And for the $millions spent on each machine, the airports could alreay have a whole pack of cute tail wagging dogs on duty. And it would take a whole lot shorter time frame to train and deploy the dogs.</font>

Exactly! onedog (and 534 other people who think like onedog) for Congress! http://www.flyertalk.com/forum/smile.gif

The beagles wouldn't take up as much room as an SUV, like the machines do, and they would certainly keep the kids (and probably some grown-ups) entertained while they sniffed. http://www.flyertalk.com/forum/smile.gif

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by L-1011:
No, I wouldn't. I will probably sleep better if I don't know how much has been wasted, eh, spent, on it. And it really doesn't matter if they spent $5 or $5,000,000,000. As long as it doesn't work, it doesn't work. But we still have to use them. Aren't political decisions nice? Especially in an election year.</font>

If you did care to find out, you would learn that the bomb detection equipment was not manufactured through government research but rather through private enterprise.

It's not your money that was spent on the R&D, so it's not a political decision at all. The incorrect assumption that it was publicly funded R&D is why the ranting is so off-base.

It's also important to realize the equipment is too good. It flags baggage with potential dangerous materials for hand-search. That's false positive.

What happens when a bag gets flagged... it's hand-searched. The exact same procedure we're talking about.

The SAFETY factor of the new bomb detection equipment is high. It's just that it causes DELAY and MORE WORK for screeners.

Facts would help the discussion, rather than operating on incorrect assumptions.

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Plato90s:
The incorrect assumption that it was publicly funded R&D is why the ranting is so off-base.</font>

High R&D cost equals high sales price equals more of my tax money is used to purchase those thingies. It doesn't matter who funded the R&D. The manufacturer will cover their costs.

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">It's also important to realize the equipment is too good. It flags baggage with potential dangerous materials for hand-search. That's false positive.</font>

To me, that's not good. It just means that the parameters aren't set correctly. A flase positive means that my bag gets opened for no reason.

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Facts would help the discussion, rather than operating on incorrect assumptions.</font>

I have the facts, it's the consequences of the operation of this equipment I don't like.

Plato90s - So do you have any idea about the cost of the lost revenue by forcing verdors out of small airports, requiring siginificant structural changes and lost space for passengers given the space the Behemoths and its support staff (read that TSA people) need - to include the tables and search area for all those false-positives?

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by bdschobel:
Does it really matter how much money has been spent? The equipment does not work -- that is well established. The error rate (false positives) is somewhere around 30 percent. What more do you need to know?</font>

For one thing, you need to know the facts.

The 30% figure is not an error rate. The machine's purpose is to separate the bags into two categories, clear and not-clear. The not-clear bags are diverted for additional checks which would be too time and labor consuming to apply to every checked bag.

This isn't unlike the metal detector through which passengers walk. The detectors selects a lot of people who aren't carrying weapons, probably a higher percentage than the bag screening machine. The purpose is to quickly eliminate a large percentage of the passengers, or the bags, from additional screening.

Same type of system is applied to the drug testing to which many airline employees are subjected. The first test is quick (and cheap) but has a high rate of false positives. Only those samples that show positive on the first test are sent on to the more involved, and accurate, tests.

If it were easy to quickly and economically scan every checked bag for explosives then we would have been doing it decades ago.

My definition of keeping my items safe does NOT include checking an unlocked bag. Even though I always use a rollaboard, the idea of making my things readily available for anyone to pillage is not acceptable.

From Douglas Adams' The Restaurant at the End of the Universe:

A junior Disaster Area accountant, visiting the shipyard where this ship was being constructed, had demanded to know of the works foreman why the hell they were fitting an extremely expensive teleport into a ship which only had one important journey to make, and that unmanned. The foreman had explained that the teleport was available at a ten per cent discount and the accountant had explained that this was immaterial; the foreman had explained that it was the finest, most powerful and sophisticated teleport that money could buy and the accountant had explained that the money did not wish to buy it; the foreman had explained that people would still need to enter and leave the ship and the accountant had explained that the ship sported a perfectly serviceable door; the foreman had explained that the accountant could go and boil his head and the accountant had explained to the foreman that the thing approaching him rapidly from his left was a knuckle sandwich. After the explanations had been concluded, work was discontinued on the teleport which subsequently passed unnoticed on the invoice as ``Sund. explns.'' at five times the price.

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Plato90s:
Maybe you'd like to know how much money has already been spent to develop the existing equipment before criticizing.

Then again, your use of the word "idiot" already tells us plenty about what you think.</font>

Wow. This machine appears then to have a 0% false negative rate, yet a 30% false positive rate (troubling math and reliability!). And even at that metric, 3 of every 10 bags (assuming no explosives that day) still have to be checked further.

From aviationplanning.com : "So here we have possibly hundreds of pieces of luggage - Samsonites, golf bags, skis, boxes, car seats, and who knows what else, piling up at peak periods in front of ten or twelve CT machines each the size of a Buick Roadmaster and which can process only about 1.5 to 2 bags a minute."

If you check a bag on Saturday, it will arrive... Tuesday? No, wait we are also doing 100% bag matching... "Please arrive at the airport 2 days before your flight is scheduled to depart".

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by LarryJ:
The 30% figure is not an error rate. The machine's purpose is to separate the bags into two categories, clear and not-clear. The not-clear bags are diverted for additional checks which would be too time and labor consuming to apply to every checked bag.</font>



------------------
"Give me Liberty or give me Death." - Patrick Henry

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Spiff:
Wow. This machine appears then to have a 0% false negative rate, yet a 30% false positive rate (troubling math and reliability!). And even at that metric, 3 of every 10 bags (assuming no explosives that day) still have to be checked further.</font>

This statement only shows you don't understand what you're talking about.

30% false positive rate is how many of the ones flagged as having potentially dangerous substances is wrong. 3 out of 10 flagged bags.

Unless you think EVERY bag is flagged for secondary screening, the actual percentage of bags which have to be hand-checked is far less than 30%.

Your failed attempt to criticize a process you don't even understand merely underscores the depth of your bias against all security measures.

Plato,

You are dead wrong. The error rate refers to the percentage of perfectly safe bags that are INCORRECTLY flagged as dangerous, requiring them to be hand-checked.

If you were correct -- and I assure you that you are not -- then 70% of the flagged bags would in fact contain explosives. You know better than that. NONE of the bags contain explosives.

Bruce

http://www.restoringamerica.org/archive/elected/homeland_insecurity.html

http://www.worldnetdaily.com/news/article.asp?ARTICLE_ID=28373

""It's time the public become aware of the fact [that] what they're watching now is show business," Slepian told WND. "What they're looking at in airports amounts to very little. The public has been misled. It has to be misled because otherwise it wouldn't travel, and the economy hinges on air travel. I don't endorse it, but that's the reason for it."

He adds, "Congress, and America, for another year, will have to cross their fingers, that al-Qaida doesn't strike again."

So, Plato, are you claiming that these machines actually have a 9% false positive rate (30% of 3 of 10 flagged bags)? Could you please cite your source?

Yeah, you're right. Maybe you could explain it a little more clearly for those of us who obviously don't understand the process. Please, feel free to use big words.

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Plato90s:
This statement only shows you don't understand what you're talking about.

30% false positive rate is how many of the ones flagged as having potentially dangerous substances is wrong. 3 out of 10 flagged bags.

Unless you think EVERY bag is flagged for secondary screening, the actual percentage of bags which have to be hand-checked is far less than 30%.

Your failed attempt to criticize a process you don't even understand merely underscores the depth of your bias against all security measures.</font>



------------------
"Give me Liberty or give me Death." - Patrick Henry

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Spiff:
And even at that metric, 3 of every 10 bags (assuming no explosives that day) still have to be checked further.</font>

Heck of a lot better than having to check 10 out of ten, isn't it? Or should we just count on the passenger's overpowering the bomb?

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">From aviationplanning.com</font>

Yes, excellent source, that is.

The machines process more than 1.5 to 2 bags per minute. The low rate was the rate when it's sitting out in the lobby separate from the rest of the baggage handling system. Once it's integrated into the system the rate increases significantly.

We're checking essentially zero bags now, and somehow life goes on. Get a grip.

Bruce

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by LarryJ:
...This isn't unlike the metal detector through which passengers walk. The detectors selects a lot of people who aren't carrying weapons...</font>

Darn, why am I always one of the (3%, 10%, 7 of 10, 9% whatever) false positives!? http://www.flyertalk.com/forum/wink.gif

You're right.

Total false-positive rate is 17-30%, so up to 3 bags per 10 screened is flagged for further inspection, which may include a hand search.

I don't agree, however, that a bomb threat is so inconsequential as to make it pointless to deploy the best scanner we can get a hold of. If the Invision one is the best, then I'm in favor of using it.

So the bottom line is that we are going to spend billions on machines that cry wolf upwards of 30% of the time...that's upwards of 300 million bags a year that will be "suspect" until searched and cleared.

And we will spend billions each year on personnel to staff these wonderful devices.

I'm not real bright, so could someone please tell me again why it's so important that we screen every bag for bombs...

My economics profs convinced me early on that we can't have everything, no matter how much good it does. What are we not spending money on because we're spending it on these devices? Couldn't we better spend this dough on fixing real problems facing Americans?

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by bdschobel:
We're checking essentially zero bags now...</font>

Shows how much you know.

OK, then, the percentage of checked bags being inspected is in the low single digits -- and nearly all of them are on international flights. That's "essentially zero" compared to what's coming in less than 2 months.

And there's no need to be so superior. If you believe that you have information to contribute, then do so. But don't merely tantalize us with your great knowledge!

Bruce

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by FWAAA:
So the bottom line is that we are going to spend billions on machines that cry wolf upwards of 30% of the time...that's upwards of 300 million bags a year that will be "suspect" until searched and cleared.

And we will spend billions each year on personnel to staff these wonderful devices.

I'm not real bright, so could someone please tell me again why it's so important that we screen every bag for bombs...

My economics profs convinced me early on that we can't have everything, no matter how much good it does. What are we not spending money on because we're spending it on these devices? Couldn't we better spend this dough on fixing real problems facing Americans?</font>

Let's think about how many billions of dollars and lost lives a single bombing could cost.

Do you buy insurance?

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Plato90s:
Let's think about how many billions of dollars and lost lives a single bombing could cost.

Do you buy insurance?</font>

Yes, I insure against risks that seem likely to occur. I don't bother insuring against risks that I conclude are too remote to be cost-effective. Probability plays a large part of the equation.

As to whether this potential risk is worth the monumental expense, we apparently disagree. http://www.flyertalk.com/forum/smile.gif

The premium for this bomb insurance looks kinda steep to me.

Does this mean you will be canceling your asteroid insurance? http://www.flyertalk.com/forum/confused.gif

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by FWAAA:
Yes, I insure against risks that seem likely to occur. I don't bother insuring against risks that I conclude are too remote to be cost-effective. Probability plays a large part of the equation.

As to whether this potential risk is worth the monumental expense, we apparently disagree. http://www.flyertalk.com/forum/smile.gif

The premium for this bomb insurance looks kinda steep to me.</font>



------------------
"Give me Liberty or give me Death." - Patrick Henry

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Spiff:
Does this mean you will be canceling your asteroid insurance? http://www.flyertalk.com/forum/confused.gif

</font>

That and my entire-universe medical policy rider (in case of alien abduction). http://www.flyertalk.com/forum/smile.gif

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by FWAAA:
Yes, I insure against risks that seem likely to occur. I don't bother insuring against risks that I conclude are too remote to be cost-effective. Probability plays a large part of the equation.</font>

There is a long history of airline bombings including such noteable cases as Pan Am 103 and the recent Al Qeada plot to bomb 12 US airliners over the Pacfic within a 24 hour period. The Al Qeada plan was only stopped because they started a fire in the lab they were using to make the explosives.

There is little doubt that US airliners will be the target of bombing attempts again in the future.

<font face="Verdana, Arial, Helvetica, sans-serif" size="2">Originally posted by Plato90s:
The SAFETY factor of the new bomb detection equipment is high. It's just that it causes DELAY and MORE WORK for screeners.</font>

I don't think the safety factor is high at all. Like it or not, there are humans operating those machines, and as long as those humans perceive the task as "delay" and "more work", they will become complacent. It is a simple fact of human nature.

When people are bombarded with too much information, they stop trying to filter the information for relevance. How many of us have signed up for one e-mail newsletter or another, thinking it would help us stay more informed of important goings-on, only to find that most of the time there was nothing important going on? You know what happens - you start ignoring the newsletter; pretty soon the e-mails are being deleted unread; you're no more informed about important happenings than you were before. (Someone please remind me what the proper psychology term is for this.) Alerts on 30% of checked bags are waaaay too much information when the screeners are really interested in the bags that contain bombs, which constitute a far smaller percentage. The consequences are predictable: people will get into "oh, it's just the stupid machine going off for the 400th time today" mode and our airports will be no safer than they are now, though we'll be poorer and a lot more annoyed. Too much information can be worse than none at all.

Imagine if we hired a team of humans to inspect checked baggage for bombs at all US airports. Let's keep things in perspective here: checked-baggage-bombs are very rare, occurring only a few times a decade worldwide. For any of these people, seeing a single actual bomb in a bag in the course of a career would be a lot. Now you've got these people inspecting bags, and 30% of the time the inspector has to tap his supervisor on the shoulder and say, "Um... There might be a bomb in this bag. Could you take a look?" Remember, one bomb in a career would be a high number, yet these people can't send off a measly 737 flight with 120 pax without alerting the supervisors to possible bombs thirty-six times. If humans were doing this we would laugh them right out of the airports. "But sir, I haven't missed a single bomb yet!" One wonders if the boss would be able to control his riotous laughter at that one. Yet when machines are doing the same thing, all of a sudden it's the answer to all our security problems.

The current generation of bag screening machines may be the best we've got, but there are times when the best isn't good enough. Quite honestly I think Miss Cleo has a better chance of finding a bomb in a bag than a cadre of Univac-esque behemoths, and at $2.95 a minute (or whatever it is) the psychic route is cheaper too. Comprehensive explosive detection is a lofty goal, but we mustn't let it become so lofty that we get our heads in the clouds. My opinion on the current technology: Interesting proof of concept; try me again in five years.

Incidentally, the European airports claim to already have a complete explosive detection program for checked baggage, yet they don't seem to be having any of our problems. One does hear about the occasional checked bag that has had to be hand-searched, occasionally without the owner present, but the infrequency of these stories indicates it can't possibly be a hundreds-of-times-per-airport-per-day occurrence. Why do things seem to be going much more smoothly over there? I'd be interested to hear from someone who has more knowledge of the subject.

[This message has been edited by Factotum (edited 11-06-2002).]