Hermes A321 at Lyon on Mar 29th 2013, runway excursion

Last Update: August 19, 2015 / 16:33:39 GMT/Zulu time

Bookmark this article
Incident Facts

Date of incident
Mar 29, 2013

Classification
Incident

Aircraft Registration
SX-BHS

Aircraft Type
Airbus A321

ICAO Type Designator
A321

The French BEA released their final report concluding the probable causes of the accident were:

Causes of the Accident

Continuing an approach below the stabilisation height with a speed significantly higher than the approach speed shows that the crew were not adequately aware of the situation, even though they mentioned several times their doubts on the marginal meteorological conditions and on the difficulties in reducing the aeroplane’s speed.

Continuing this unstabilised approach at an excessive approach speed triggered, below 150 ft, an uncommanded increase in engine thrust. The crew’s delayed A/THR reduction below 20 ft made it impossible for the aeroplane to slow down sufficiently for about 15 seconds after passing the threshold.

After descending through 20 ft, the copilot’s inappropriate flare technique and the dual input phenomenon caused by the Captain significantly lengthened the flare phase. The remaining runway distance after the touchdown made it impossible for the aeroplane to stop before the end of the runway.

The following factors contributed to continuing the unstabilised approach and the long flare:

ˆˆ- a flight duty period of nearly 15 hours which likely led to crew fatigue;

- incomplete preparation of the approach which meant the crew was not aware of the risks on the day (tailwind, wet runway);

- the non-application of ATC procedures that require controllers to ensure aircraft are provided with localiser interception at the latest 10 NM from the runway threshold, with a maximum convergence of 30° and a maximum speed of 160 kt;

ˆˆ- partial application of standard procedures (SOP), impaired task sharing and degraded CRM, which meant the crew was unable to manage optimally the aeroplane’s deceleration. These factors contributed to a progressive deterioration in situational awareness that meant that they could not envisage rejecting the approach and landing;

ˆˆ- the A/THR anomaly which maintained the aeroplane at a high energy level during the landing phase;

- an inadequate procedure for taking over the controls that led to the dual input phenomenon.

The following organisational factors contributed to the crew’s poor performance:

- the choice of flight crew recruitment profiles by the operator, motivated by economic considerations, and inadequate airline conversion, led to operating aeroplanes with crews that were relatively inexperienced on type and in their roles as captain or copilot;

ˆˆ- improper and inappropriate application of the regulatory provisions that allow an extension of flight duty time in case of “unexpected circumstances” without taking into account the predictable risk of excessive fatigue for the crew;

ˆˆ- the absence of suitable initial oversight which made it impossible for the HCAA to focus on the predictable potential operational weaknesses of Hermes Airlines.

The BEA reported that the first officer (26, ATPL, 600 hours total, 314 on type) was pilot flying, the captain (44, ATPL, 7,096 hours total, 1,341 hours on type) was pilot monitoring. The crew briefed for a CAT I ILS landing on runway 36R (landing distance available 2640 meters/8660 feet) and computed a Vapp of 141 knots. ATC cleared the aircraft to descend to 4000 feet, speed 220 KIAS. The first officer, on advice by the captain, used speed brakes to accelerate the descent, at 220 KIAS flaps one were selected. The aircraft incepted the localizer 12.5nm before the runway threshold, at 10nm before the runway threshold the aircraft was still at 220 KIAS, 251 knots over ground at a rate of descent in excess of 2000 fpm. About 9nm before runway threshold the aircraft intercepted the glideslope at 217 KIAS and 1500 fpm rate of descent.

Descending through 2000 feet AGL the aircraft slowed through 205 KIAS, flaps 2 were selected at 1550 feet AGL at 203 KIAS, winds were reported from 130 degrees at 6 knots, a speed of 180 KIAS selected into the speed window, the landing gear was extended.

The aircraft descended through 1000 feet AGL at a speed of 198 KIAS, 57 knots above approach speed outside the window for a stabilized approach, speed selected into authotrust was still 180 KIAS.

The captain instructed the first officer to use managed speed mode, the first officer selected managed speed which reduced the speed selection to 153 KIAS.

At 850 feet AGL the aircraft slowed through 193 KIAS, flaps 3 were selected and at 625 feet AGL FULL configuration was selected at a speed of 184 KIAS.

The aircraft descended through 500 feet AGL at 179 KIAS, 4 degrees nose down, more than 1100 fpm rate of descent. The autopilot was disengaged at 200 feet AGL, the autothrust was kept engaged.

Descending through 140 feet AGL at a speed of 158 KIAS the autothrust system started to increase engine thrust, N1 increased from 30 to 54% and the aircraft began to accelerate.

The aircraft crossed the runway threshold at 60 feet AGL at 160 KIAS, 7 knots tail wind component.

About 3 seconds after descending through 30 feet AGL the thrust levers were pulled into the idle detent. The first officer kept low amplitude nose up inputs, the pitch changed from about 1.4 degrees nose down to 1.7 degrees nose up, the rate of descent was about 600 fpm, the aircraft descended through 21 feet about 500 meters/1650 feet down the runway, the engines slowed through 69% N1, the airspeed was 163 KIAS. The captain began to provide nose down inputs, the first officer maintained nose up inputs, in sum nose up, the automatic call "DUAL INPUT" occurred.

The aircraft touched down 1600 meters/5250 feet past the runway threshold at 154 knots over ground, the spoilers deployed and maximum reverse thrust was selected, the crew applied asymmetric braking. The aircraft began to slow at -0.4G, but overran the end of the runway at about 75 knots over ground and came to a stop approximately 300 meters past the runway end.

The aircraft sustained damage to both engines and minor damage to the landing gear.

The BEA analysed: "When low visibility conditions prevail at Lyon, the ATC procedure requires the controller to have the localizer intercepted at 160 kt and 10 NM from 36R threshold at the latest. On the day of the event, the controller did not apply this instruction and the aeroplane intercepted the localizer at about 12 NM at a speed of 220 kt. Air traffic control explained that this speed constraint is only useful for ensuring aircraft spacing and landing rate. In practice, it is not taken into account when there is little traffic. It is the crew’s responsibility to manage the speed of its aeroplane. Nevertheless, the application of the ATC speed regulation procedure by the controller would have provided the crew with the opportunity to anticipate speed reduction during approach. In September 2013, the DGAC had drawn operators’ and air traffic control service providers’ attention to the risks related to excessive speed on final. It recommended a maximum speed of 180 kt, reducing, 8 NM from the runway threshold. After the accident a new method of speed management was implemented at Lyon."

The BEA continued, pointing out English as generic language on ATC would be useful: "During the radar vectoring of the previous aeroplane (A319 Air France flight AF-DD), the controller shared his doubts with the crew on the aeroplane’s high ground speed (250 kt). The crew then answered that they were going to anticipate landing gear extension. Four minutes later, the crew of SX-BHS were established on long final in the same conditions (4,000 ft / 250 kt). Unlike the previous flight, the controller did not share his doubts, explaining that the aircraft was the same type as the previous one and that their performance should be identical. The controller’s initiative of sharing his doubts with the Air France crew may have helped them to raise their awareness of the deceleration difficulties associated with their high ground speed. However, this radio communication in French could not be understood by the crew of SX-BHS. They were thus deprived of any chance of becoming aware of the difficulties on deceleration."

With respect to continuing the approach at and below 1000 feet AGL the BEA analysed: "Standard procedures (SOPs) require the PM to monitor the flight parameters in order to ensure the approach is stabilised at a height of 1,000 ft AAL in IMC conditions. When significant deviations are detected, it is expected that the crew perform a missed approach. In this particular case, the tacit decision to continue the approach indicates that the crew were apparently unaware of the risks incurred or that he does not feel able to make a missed approach. Testimony indicates that they never thought nor mentioned a go-around, except during the initial briefing."

The BEA continued analysis: "The investigation showed that taking over the controls leading to dual input occurs more frequently during the final approach phase or during flare when the copilot is PF. In many cases, the copilot is on line flying under supervision. It therefore appears that the scenarios for taking over the controls during training sessions are not in line with the most frequently encountered situations in operation. The remaining runway distance after touchdown proved to be insufficient to allow the aeroplane to stop on the runway despite energetic braking by the crew."

With respect to balked landings the BEA analysed: "At the time of the accident, there was no procedure for rejected landings in the manufacturer’s FCOM. This aspect was mentioned in the FCTM and reminded crews that they could perform a go around as long as the thrust reversers were not deployed. The manufacturer considered that this situation was covered by the association of the FCOM “Go Around” procedure and the specific information provided in the FCTM."

The BEA analysed with respect to crew performance: "The crew’s flight duty period was close to 15 hours at the time of the event. Observation of their performance showed alterations which were symptomatic of fatigue. ... The day before the event flight, the Operations department of Air Méditerranée had advised Hermes Airlines to provide for an augmented crew because of possible extension of the flight time due to a possible technical stop. This stopover was therefore foreseeable and the use of the waiver for unforeseen circumstances was thus highly debatable."

With respect to Hermes Airlines the BEA thundered in analysis:

According to the Captain, on the day before the flight, he had refused a crew changeover on the outbound leg as this solution seemed to him to be a last-minute fix.

However, the investigation showed that he had had to handle, without any controlled rest, a flight situation requiring sustained attention in the following domains:

-ˆˆ supervision of a young inexperienced copilot, a situation similar to line training;
-ˆˆ performance of a flight at the aeroplane’s limit of endurance, requiring meticulous monitoring of fuel during the outbound leg;
-ˆˆ managing a stopover in Dakar with a delay and programming a technical stop in Agadir, increasing the duty period;
-ˆˆ arrival in Lyon, at night, in deteriorated meteorological conditions.

Refusal of extra payload from Dakar to avoid a technical stop in Agadir would in addition have incurred an increase in operating costs for the airline which the Captain was afraid he would be blamed for. Interviews with Hermes Airlines personnel indicated that they were concerned with limiting costs to a minimum. It seemed that some even feared losing their jobs in the event of an error incurring substantial additional costs. The decision of the manager in charge to dismiss the Captain after the accident was also likely to increase the employees’ perception of this risk. The Captain’s decisions were made in a context of adverse economic pressure.

Hermes Airlines’ management seemed to accept, or indeed even favour this technique of applying a waiver for unforeseen circumstances that allow an extension of the flight duty period to 15 hours, in order to avoid resorting to augmented crews, a more expensive solution.
Incident Facts

Date of incident
Mar 29, 2013

Classification
Incident

Aircraft Registration
SX-BHS

Aircraft Type
Airbus A321

ICAO Type Designator
A321

This article is published under license from Avherald.com. © of text by Avherald.com.
Article source

You can read 2 more free articles without a subscription.

Subscribe now and continue reading without any limits!

Are you a subscriber? Login
Subscribe

Read unlimited articles and receive our daily update briefing. Gain better insights into what is happening in commercial aviation safety.

Send tip

Support AeroInside by sending a small tip amount.

Related articles

Newest articles

Subscribe today

Are you researching aviation incidents? Get access to AeroInside Insights, unlimited read access and receive the daily newsletter.

Pick your plan and subscribe

Partner

Blockaviation logo

A new way to document and demonstrate airworthiness compliance and aircraft value. Find out more.

ELITE Logo

ELITE Simulation Solutions is a leading global provider of Flight Simulation Training Devices, IFR training software as well as flight controls and related services. Find out more.

Blue Altitude Logo

Your regulation partner, specialists in aviation safety and compliance; providing training, auditing, and consultancy services. Find out more.

AeroInside Blog
Popular aircraft
Airbus A320
Boeing 737-800
Boeing 737-800 MAX
Popular airlines
American Airlines
United
Delta
Air Canada
Lufthansa
British Airways