Original subtitles

Zurich airport,

November 2001.

Crossair flight 3597 is one of the few planes

left to land.

The commander was flying out of Zurich

for the last 20 years.

He was really used to this airport.

The lone controller in the tower

tracks the approach.

Approach runway 28 for you.

I have ground contact.

We're continuing on.

Crossair 3597,

continue speed reduction to final approach speed.

But for some reason

the crew can't find the runway.

Damn.

He said he saw the runway 1.3 miles.

And the flight ends in disaster.

Crossair 3597,

this is Zurich tower.

Do you copy?

We have a possible emergency...

He collided with a hill,

so obviously he was too low.

A terrible mistake sent a passenger jet

dangerously off course,

killing 24 people.

The clues will lead investigators to a trained professional

with an astonishing past.

Mayday mayday

Crossair flight 3597

is cruising at 27,000 feet above Germany.

As the autopilot flies the plane,

the crew begins reviewing procedures for their landing.

According to the report,

the apron and taxi ways are wet.

Breaking action's not specified.

So the friction report is missing?

Indeed.

That usually means they haven't been out to check lately.

Indeed.

It's the last flight of the day

on this busy European route.

The 660 kilometer journey

from Berlin south to Zurich

takes only an hour and a half.

There are many empty seats on board.

A group of 21 passengers never showed up for the flight.

Internet entrepreneur, Peter Hogenkamp,

is returning from a business trip in Germany

with his partner, Jacqueline Badran.

On that day, we wanted to take the train,

but we had to be in Zurich

for the opening ceremony of the trade show,

so we decided okay,

so we need to go by plane.

I have got to get some sleep.

It had been a very busy day,

so we were very tired when we arrived at the airport.

Among the passengers

is the girl band Passion Fruit,

Germany's answer to The Spice Girls.

They're on their way to perform the last show

of a concert tour.

I was shocked when suddenly all the Passion Fruits,

all the three of them are coming over,

and they were sitting right in front of us.

Just what I need.

Flight 3597 is an Avro 146 airliner.

It's a British-made jet popular

with regional airlines like Crossair.

Captain Hans Ulrich Lutz is in command.

He's been with the airline for more than 22 years,

having served as a pilot and as an instructor.

So, tell me about the runway

and the conditions.

Overcast with light snowfall,

visibility...

First Officer Stefan Lohrer

is a newcomer to the airline.

He's had his pilot's license for just a year and a half.

He was hired by Crossair right out of flight school.

The Passion Fruits were being quite noisy.

They were excited,

I think they went from one gig to the other.

So they were still in their stage clothes.

I said oh my god,

do I really need this?

Can't I just sleep?

It's probably okay to move.

We decided to move back a few rows

and change the side where we were sitting,

so we were in a very quiet area then.

Crossair 3597,

clear to flight level 160.

The flight is slightly behind schedule.

It's due to land in Zurich in less than half an hour.

According to the latest information,

we'll be using runway 14.

What kind of approach?

ILS approach.

Okay.

An ILS, or instrument landing system approach,

means the pilots will be guided to the runway

by radio signals beamed from the airport.

And a decision altitude of 1602 feet.

Call me 100 above minimums.

Okay, I'll call 100 above minimums.

The plane's sophisticated computers

will do most of the work getting them down.

At Zurich airport, the weather is getting worse.

It's snowing,

and visibility is dropping.

Crossair 3191,

turn left heading 150.

In the tower,

controllers have started a nightly routine.

A supervisor sends out a message

notifying crews that runway 14

is being taken out of service

because of a controversial new law.

Planes landing on runway 14 must fly over

Germany on their approach.

But people in Germany have long complained

about the noise.

Because of the political issue,

airplanes had to fly over Switzerland

in their final descent to the airport of Zurich

for noise reasons.

There was no other operational

or weather reasons whatsoever.

It was purely political.

Switzerland has agreed to shut down

runway 14 after 10 pm

to keep landing planes out of German airspace,

and use runway 28 instead.

The crew hasn't received this information yet.

But First Officer Lohrer is familiar with the new rule.

Should I ask if it's still runway 14?

It's almost 10 o'clock.

Yes, I'm pretty sure it's still 14.

Zurich approach.

Crossair 3597.

Please confirm the approach in use

is the ILS 14.

Crossair 3597, you're identified.

It will be a V-O-R-D-M-E.

Approach runway 28 for you.

Hell.

What next?

Okay, fine.

Unlike runway 14,

with its sophisticated instrument landing system,

runway 28 is not equipped for ILS landings.

Instead it has a less accurate navigational system.

It's called VORDME.

A radio beacon sends information to an aircraft

about its distance from the runway,

and whether it's left or right of it.

It provides no guidance on the planes altitude.

Pilots would rather not make VOR approaches,

because the place a greater workload on the flight crew.

He would always prefer an ILS approach

because he gets guidance on the approach slope.

He can also use his autopilot to guide him down

literally on ground to the runway.

Because of its less precise technology,

runway 28 was rarely used for landings

before the new noise law.

Okay then, re briefing for runway 28.

That will be chart 13-2.

Now Captain Lutz abandons

the ILS approach he was planning,

and prepares his first officer

for the new non precision approach.

Are you familiar with the 28 approach?

Yes, I've done it a couple of times.

The crew goes over every detail

of how they will approach Zurich airport.

East 6000 feet.

There are only a few planes

left to land tonight. - 3597.

Including flight 3597

and two other Crossair flights just ahead of it.

As the aircraft approach the runway,

controllers need to keep them apart.

Crossair 3597,

reduce speed to 180 or less.

Speed 180 or less, Crossair 3597.

Captain Lutz makes the final turn

to line his jet up with the runway.

160 knots.

The plane is now in the thick

of the heavy weather surrounding Zurich.

At the airport,

the second last Crossair flight

touches down on runway 28.

Now with only flight 3597 to land,

the supervisor decides to head home early.

The controller handling the aircraft

is the only controller left in the tower.

Crossair 3597,

continue speed reduction to final approach speed.

Zurich tower this is Crossair 3891.

She gets a report from the plane

that just landed.

Visibility approaching 28 was borderline.

We could only see the runway from 1.3 miles out.

But Captain Lutz knows Zurich airport well.

He continues towards the runway.

Gear down.

The plane begins the final stage

of its approach.

Cab and crew,

prepare for landing.

Captain Lutz begins slowing the jet,

and puts it into a steep descent.

Six miles is checked.

Yes.

Flaps three three.

Flaps three three selected.

Final check.

Confirm three greens.

Is checked.

116 knots.

The descent gets even steeper.

Full flaps set.

Checked.

Flight 3597 is about three minutes

from the runway.

Captain Lutz expects it to come into view any second now.

100 above.

Do we have ground contact?

Yes.

Passengers prepare themselves for landing.

We were pretty close to the airport,

so we think okay, this is going smoothly.

But it was dark and it was snowing,

so we couldn't see anything.

In the tower,

the controller monitors the flight

to make sure it's lined up with the runway.

Crossair 3597, clear to land runway 28.

Two four the minimum.

Two four.

The flight has hit decision height

2400 feet above sea level,

1000 feet above the ground near Zurich.

Decision height is the minimum height

you can descend before you need to decide

whether you can see the runway

and continue the approach visually,

or you have to make a decision to go around.

I have ground contact.

We're continuing on.

500.

Captain Lutz still can't see the runway lights,

but knows that the flight ahead of him

had the same problem.

Damn.

He said he saw the runway 1.3 miles.

They should almost be

on top of the runway by now.

But for some reason they still can't see it.

Make it go around?

Yeah.

Crossair 3597...

The crew decides to abort the landing.

Go around.

And pushes the plane to its limits

in an effort to get it to climb.

Captain Lutz throttles up his plane's four jet engines

as he aborts his landing at Zurich airport.

But it's too late.

There's not enough time for the engines

to get to full power and clear a tree-covered hill

suddenly looming dead ahead.

Residents of a nearby village

witness flight 3597's final moments.

I was walking down with my dog.

All of a sudden,

I saw that aircraft coming from behind

the forest, the little hill.

I thought oh, it's very low,

and it's not on the same route

as the aircraft normally are

coming to land the airport of Zurich.

The jet clips the treetops.

Suddenly, it was shaking very hard.

What's happening, something is wrong.

Moments later,

the aircraft plunges into the hillside.

Damn!

Crossair 3597,

this is Zurich tower.

Do you copy?

Crossair 3597,

this is Zurich tower.

Do you copy?

All of a sudden,

the whole sky in the west was getting really orange,

like a big fire.

Like an explosion.

The impact tears the aircraft apart.

The wings broke off,

and the fuel is in the wings,

so the flames really sort of shot inside the plane.

Flight 3597,

this is Zurich tower.

Do you copy?

This is Zurich tower.

We have a possible emergency in approach sector east.

I repeat, approach sector east.

Crossair flight 3597 has crashed

near the village of Bassersdorf,

northeast of Zurich.

Some people have survived the accident,

including Peter Hogenkamp

and Jacqueline Badran.

I don't have any memory of how we really walked out.

Because the back was cut off,

and we were just able to walk out there.

One member of Passion Fruit also survives.

But her two band mates seated just one row ahead of her

have been killed.

When they give you these instructions

in the beginning when you get on the plane,

how come they never tell you how far

you have to run away from a burning plane.

Because they never tell you.

Let's get away from here.

So we thought, I don't know 100 meters.

Within minutes,

emergency workers get to the crash site.

Hey.

Over here.

Over here!

Over here!

24 of the 33 people

on board flight 3597 are dead,

including Captain Lutz

and First Officer Stefan Lohrer.

Daniel Knecht,

an investigator from Switzerland's

aircraft investigation bureau,

arrives at the scene.

I got the call from our rescue service

about half past ten in the evening.

Half an hour later, I was on the accident site.

The rear part of the fuselage was quite intact.

It was also the region where

most of the survivors step out without injury.

For Peter Hogenkamp and his partner,

it appears that moving to the rear of the plane

saved their lives.

I think it was just good luck.

I think it would've been a whole different story

if we wouldn't have changed seats.

So the Passion Fruits may have saved our lives.

While rescuers care for the survivors,

Knecht and his team survey the crash site.

The location of the wreckage far from the airport

is of interest to the investigators.

4 kilometers and 50 meters from the runway.

The aircraft was not where it was intended to be.

He collided with a hill east of runway 28,

so obviously he was too low.

The first question was,

why was this aircraft too low?

The black box flight recorders

are found in the wreckage.

They're in good condition.

Investigators are well aware

that this is the worst aviation disaster

on Swiss territory in over a decade.

They're under pressure to find answers fast.

While work continues at the crash,

a command post is established nearby.

Jean Overney heads the Swiss investigation bureau.

He takes charge of the Crossair file.

The public was very anxious about this crash.

The media were raising many questions,

so we naturally had to open a more in-depth investigation.

The accident happened at the end

of a long day for Captain Lutz.

But given his background,

investigators must consider other causes.

We knew that the captain was very experienced.

He had about 17,000 flight hours during his whole career.

Flight instructor, flight examiner.

Within the company for more than 20 years.

The flight recorders

are sent to a lab in Paris to extract their data.

At the same time,

investigators have begun retrieving fragments

of wreckage from the crash site.

They look for tell-tale signs of trouble.

The position of switches is very important.

Also, if you have some indication panels,

for example, with bulbs in it,

after a crash you can analyze

the different wires in the bulbs

and then you can determine which indications

or even warnings were active at the moment of the impact.

An investigator comes across

an intriguing piece of wreckage.

Oil gauges provide information about the oil

flowing to the plane's four turbo fan engines.

Curiously, one of them is installed upside down.

This raised a lot of questions for us

and for the media.

Everyone wanted to know how you could fly a plane

with an instrument installed upside down.

It wasn't a very important instrument.

You could still read it,

even if it was installed upside down.

But if this obvious problem was never fixed,

what else could be wrong with the plane?

Obviously,

there were maintenance issues here.

We looked at how the company

was doing the maintenance of the planes.

While this review is being launched,

investigators study the plane's engines

to determine if they were running

at the time of the crash.

Soil, tree branches, and other debris

lodged inside the engines suggest they were.

We checked them out visually,

and we saw that the four engines were turning

in the moment of the accident.

Authorities need to be sure.

They will have to check their findings

against data from the flight recorders.

In the meantime,

the location of the crash suggests something

may have tragically mislead the crew

on their descent towards Zurich airport.

Knecht focuses on the flight charts

they used to plan their approach.

Is the problem with the approach

or is the problem with the aircraft?

This kind of question was dominant

when we started the investigation.

Are you familiar with the 28 approach?

Yes, I've done it a couple of times.

The approach chart provides pilots

with directions to the runway.

Trasadingen Zurich.

Details about radio frequencies

as well as altitudes and speed

for various stages of the landing.

When investigators study an approach chart

like the one the crew used,

they make a shocking discovery.

The chart does not show the hill that the plane hit.

An approach chart is a very important thing

because an indication which is not correct

on this approach chart could mislead a crew,

and so easily make them do something wrong.

So, tell me about the runway and the conditions.

Overcast with light snowfall.

A faulty chart would certainly

have been a danger to a crew unfamiliar

with Zurich airport.

But Lutz had flown in and out of Zurich countless times.

He and his first officer were both based in the city.

The commander was flying out of Zurich,

I think for the last 15 to 20 years,

so he was really used to this airport.

Investigators conclude

that the approach chart, while faulty,

could have tripped up some crews,

but not this one.

Some definitive news finally arrives.

The engine data from the black box

confirms the observations at the crash site.

We could see quite easily that the engines

worked absolutely normally.

Another potential cause

of the accident is eliminated from the list.

Mechanical failure.

Investigators now consider a different possibility.

I was covering the approach west position

on the night of the accident.

Could air traffic control

have somehow mishandled the plane?

Investigators study records

from the night of the accident,

and make a disconcerting find.

Zurich airport's control tower

was understaffed at the time of the accident.

He left before his shift ended.

The supervisor in the tower

went home early,

leaving a lone controller to guide flight 3597

in for its landing.

She was relatively young,

and let's say she didn't have that much experience.

Investigators now study transcripts

of conversations between that air traffic controller

and the flight crew,

looking for signs she made a mistake.

Crossair 3597, you're identified.

Speed 180 or less.

- Crossair 3597 - Zurich approach

Crossair 3597.

Clear to land.

They find no indication

the controller gave the crew faulty instructions.

But they do come across the radio call

by the pilot of the Crossair flight

that landed minutes earlier.

Zurich tower this is Crossair 3891.

Visibility approaching 28 was borderline.

We could only see the runway from 1.3 miles out.

That pilot warned of poor conditions

on the approach to runway 28.

She had options.

They wonder why the controller

didn't shut the runway down.

Because of poor visibility,

she could've reopened runway 14

in spite of the noise by law.

This would've allowed the crews

to make an instrument approach.

But she didn't do that.

Thank you, Crossair 3891.

Have a good evening.

It would have been much easier to fly ILS approach

in this kind of weather,

regardless of this political issue.

It's suspected that

without her supervisor present,

the controller lacked the experience

to make such an important call.

So it's clear that to make the decision

not to allow an approach on runway 28,

and instead make it on runway 14

demanded a certain kind of confidence,

a certain kind of courage.

However, Captain Lutz also heard

the same warning.

28 was borderline.

We could only see the runway from 1.3 miles out.

Gear down.

He kept going in spite of it.

Investigators are now certain

that the controller was in no way responsible.

The pilot can always say no.

He can always recheck the clearance.

If he thinks that landing is probably not possible,

he's free to retract an approach clearance,

and ask for a different one.

Jean Overney can't understand

why Captain Lutz didn't do exactly that.

He should've known that he didn't have to make

a non precision approach in bad weather.

When we saw the mistake the pilot had made,

we immediately asked ourselves,

well, how is it possible?

Because for an experienced pilot who was an instructor,

this was a really basic mistake.

Overney is beginning to wonder

if something in Lutz's long career

is hiding clues about his behavior

on the night of the accident.

One by one,

investigators are ruling out the possible causes

of the crash of Crossair flight 3597.

They hope the answer lies in the cockpit voice recorder,

or CVR, of the doomed jet.

After waiting many days for its data

to be sent back from France,

they now paw over it carefully,

listening for any clues that might explain

why the plane was flying so close to the ground

so far from the airport.

L nav is engaged.

Authorities get more than they expected.

Not only is the CVR complete,

but Captain Lutz offers a running commentary

on everything he does.

And on inbound track 275.

Speed is checked, flaps 18.

It's practically a road map

to understanding what went on in the cockpit.

That's not usual that the pilot flying

is talking so much about what he is doing

or what he is seeing.

For the investigation this helped a lot.

100 above.

Do we have ground contact?

Yes.

Investigators learn

that the crew of flight 3597

was looking for the runway,

and believed it would soon come into view.

From the CVR,

we saw that the commander and also his co-pilot

were not feeling some fear or feeling

that something was not going as they planned.

But what they didn't know,

was that the plane was actually

four miles shy of its target.

Damn.

He said he saw the runway 1.3 miles.

300.

Captain Lutz obviously thought his jet

was closer to the runway than it was.

Investigators consider the possibility

he was mislead by his distance measuring equipment,

or DME.

It's supposed to tell pilots how far they are

from the airport.

The DME indication would have given this distance,

and then he would have detected immediately

that he was below the approach (mumbles).

Perhaps the DME from flight 3597 was faulty.

It's been recovered from the wreckage,

but it's impossible to tell what information

it gave the pilots.

However, there is another way to check

whether the instrument was functioning or not.

The DME distance is not recorded on the FDR.

But on two occasions,

the pilots mentioned the distance.

Six miles is checked.

Yes.

But was Captain Lutz really six miles out

from the runway when he made that report?

Investigators compare the time Lutz checked his DME

with radar data on the aircraft's location

at that exact same time.

They can see that when Lutz's instruments

were telling him he was six miles out,

that's precisely where he was.

We knew that the pilots had an indication

for the distance available at that time.

An instrument problem

can now be crossed off the list

as a factor in the crash,

which only heightens the mystery.

I have ground contact.

We're continuing on.

If the instrument was working,

why did Lutz think he was closer to the runway

than he actually was?

The answer lies not in what investigators

hear on the cockpit voice recorder,

but what they don't hear.

I don't see anything after six miles.

Do we have ground contact?

Neither pilot read out a DME check

after the six mile reading.

Yes.

It tells investigators

that Lutz wasn't monitoring his vital instrument

during the final minutes of his approach.

Instead, he was preoccupied with spotting

the lights of the runway.

That's not what pilots are supposed to do.

The pilot flying

is supposed to keep his eyes on the instruments,

because all the information he needs is there.

He only switches to flying by sight

when the non-flying pilot says,

runway in sight.

Only then can the captain take his eyes

off the instruments and fly by sight.

Because the landing is performed visually.

If Lutz had checked his instruments,

he would've known how far from the runway

he actually was.

But how had he ended up such a long distance from it?

Confirm three greens.

Is checked.

116 knots.

Further analysis of the black box data

reveals that Lutz put his plane into a steep descent

as it began its final approach to Zurich airport.

But it's readily apparent his approach profile

was out of sync with the trajectory

laid out in his chart.

The approach profile for runway 28

calls for pilots to make a gradual and steady descent,

which would bring them to 2400 feet

after they've cleared the hills around Zurich.

Then pilots have to level off

until the spot the runway.

When investigators compare the approach

Lutz made with the approach profile

he was supposed to follow,

they find that his overly steep descent

brought him to 2400 feet

well before he cleared the hills.

Then he continued to descend

until he hit the ground.

It's beginning to look like Lutz

had violated one of the cardinal rules of flying.

Aircraft making non precision approaches

are supposed to level off

at minimum descent altitude, or safe height,

until they have the runway in view.

This safe height is intended to keep the aircraft

above any potential hazards.

It's really a basic error.

Every pilot is trained to respect such minimums.

Not only with VORDME approaches,

but also with ILS approaches.

But Captain Lutz kept his aircraft

in a steep descent without even pausing

at 2400 feet.

The logical next question for investigators,

could he see the runway or not?

I have ground contact.

We're continuing on.

What does he mean by ground contact?

They need to know if it was even possible

for Lutz to see the runway when he reported ground contact.

They retrace his steps in a simulator

programmed with the weather he was facing that night.

Six miles from runway.

Altitude 3300 feet.

4.8 miles.

2400 feet.

Still no runway in sight.

No, he couldn't see the runway

because there was a hill between him and the runway.

And at that altitude,

there's no way he could see it.

The conclusion,

Captain Lutz had clearly violated

his minimum descent altitude.

Whether there is flat terrain or hills,

minimums are set accordingly,

and in no way you can just bust the minimums

whether there's hills or not.

This discovery raises questions

not just about Lutz's actions,

but his young first officer's as well.

Two four the minimum.

Two four.

We're continuing on.

A first officer should not let his captain

breach minimum descent altitudes.

The co-pilot could've said no, stop,

we can't see the runway.

We're not descending.

He considered his captain

an old pilot with a lot of experience.

Not making any mistakes.

He was trusting him to much.

Also by being submissive.

Maybe he won't speak up on time

just to keep harmony in the cockpit.

There were no mechanical failures

or instrument problems with the plane,

nor was the air traffic controller negligent in any way.

But it is now clear that Captan Lutz

made a series of catastrophic piloting errors

that doomed flight 3597

and claimed the lives of 24 people.

Investigators dig deeper

into Captain Hans Ulrich Lutz'z flying record

to see if there's anything in his long background

that could explain why he crashed flight 3597.

What they discover is truly alarming.

Lutz applied to flight school when he was 17 years old.

He was rejected three times

because of his lack of education.

He finally got his license at the age of 20.

But continually failed exams to upgrade

due to his inadequate comprehension of navigation systems.

Captain Lutz may have had a long career,

but it was not a stellar one.

Years of flying and flying hours

don't say anything about competence.

For some reason though,

his record as a below average pilot

didn't prevent Crossair from hiring him in 1979.

There was a red line through his career

where several incidents happened before

and he had missed several checks.

He had some limits.

What's hard to understand for me

is that they were identified

and the decision was made to keep him flying.

Captain Lutz's limited abilities

continued in his new job.

While commanding a sight seeing tour

of the Swiss Alps,

he made a navigation error,

and ended up in Italy.

He only realized his mistake

when his passengers spotted road signs in Italian.

In another incident,

he wrecked a 1.5 million dollar aircraft

after inadvertently retracting the landing gear

while the plane was still one the tarmac.

Lutz's failures forced Crossair

to fire him as in instructor.

But they allowed him to continue flying passenger planes.

I personally do pilot assessments.

And he would never have gone through my assessment

with a positive recommendation.

Okay.

Approach course 1, 3...

Investigators wonder why Crossair

allowed Lutz to stay on as a pilot

if he was not entirely competent.

The answer appears to be they may have had little choice.

Investigators' first glimpse

into the problems at Crossair

came after they found the upside down gauge

at the crash site.

We had to check if the employees

had received the proper training,

if they had the qualifications required

for this kind of work.

No glaring maintenance issues were found.

But a closer look at Crossair's hiring practices

explains why Captain Lutz was allowed to keep flying.

The airline may have grown too fast

throughout the 80s and 90s.

The company expanded very rapidly,

and with that success,

it's clear that you have to quickly increase the staff.

Office workers, mechanics,

that's one thing.

But pilots, you have to go out and find them.

In those days,

they were desperate for pilots.

And somehow no decision was made to lay him off

because of his questionable performance

during his career.

Captain Lutz's failings

caught up with him on November the 24th 2001.

Make it go around?

Go around.

24 people paid with their lives.

After the accident report was issued,

sweeping changes were made.

Swiss aviation authorities had Crossair

increase the number of employees overseeing pilots.

They also scrutinized other airlines

to determine if there were other

substandard performers like Lutz.

One thing is for sure,

that selection process has improved considerably.

Of course, awareness of people's performance

during their career was increased considerably.

Crossair went out of existence in 2002.

It was folded into the Swiss National Carrier.

Once Crossair became Swiss,

they adopted the previous Swiss Air selection process.

And I can say today,

which is already nine years since,

that safety has improved considerably,

and also quality of the pilots

that are hired by companies like Swiss.

Since the crash of flight 3597,

Zurich's runway 28 has been upgraded

with both an instrument landing system

and an automatic warning that alerts controllers

when a plane's approach is too low.

There has never been another accident

involving that runway.

But the Crossair tragedy,

and the circumstances that spawned it

isn't an isolated case.

Regional airlines have boomed worldwide since the 1980s,

particularly in the United States,

where they account for half of all flights

and carry 100 million passengers a year.

In the rush to fulfill consumer demand

for cheap air travel,

these airlines often hire less experienced pilots,

and pay them far lower salaries

than their counterparts at major carriers.

Some even neglect to test their pilots for basic skills.

There is no such thing as compulsory pilot assessment

before they get hired.

Does not exist.

Colgan Air in February of 2009,

one of its planes crashed in Buffalo, New York,

killing 50 people.

The accident was blamed on the actions

of an inexperienced crew,

an eerie reminder of the Crossair disaster.

It underlined the urgency of improving oversight

of smaller carriers.

If you're making a product

for a large company,

you make a mistake,

you may lose part of the market.

You may lose some clients.

But an airline company can lose a plane,

its crew, and passengers.

The stakes are very different.