reducing "Crew-caused"
approach and landing
accidents 

Pilot-in-charge Monitored Approach

2013 Super Puma CFIT Sumburgh UK

Brief account : 

During a non-precision IMC approach with the Captain as PF, the helicopter descended below the MDA without either pilot having adequate visual references. Inadequate instrument monitoring resulted in a decreasing airspeed and increased descent rate going unnoticed until it was too late to effect a recovery before the aircraft hit the sea, resulting in the death of 4 passengers.      

Crew-related factors : 

The helicopter was being flown by the Captain using a non-optimal autopilot mode. "Despite the poorer than forecast weather conditions at Sumburgh Airport, the commander had not altered his expectation of being able to land from a Non-Precision Approach". The crew procedures "required the PF to fly the approach until the PNF had the required visual reference to complete the landing. The PNF was required to monitor the approach and look outside to
acquire the visual references."

The instrument approach procedure was effectively a "dive and drive" approach which may be safer at helicopter speeds than for fixed-wing aircraft. The Captain had elected to use a reducing speed technique which added to the monitoring workload. The report was critical of numerous aspects of the company's SOPs. 

Both pilots had received CRM training, but the report noted a significant experience and authority gradient between the two pilots. "[the co-pilot] accepted the commander’s ambiguous comments during the approach briefing..... The co-pilot tended to defer to the commander’s decisions, rather than questioning them ..... because of the commander’s greater experience." 

Plan continuation bias appeared to be a significant factor. Prior to the approach the Captain's comments indicated that he expected to landing and did not have any alternative plan. ".. the commander may have retained an expectation of being able to see the runway during the latter stages of the approach and did not adjust his mental model to allow for a possible level-off at MDA, or flying a go-around."  

The crew did not achieve a shared understanding of how the approach would be conducted and for a significant period (around 30 seconds at least it appears that the attention of both pilots was outside the cockpit attempting to acquire visual references, resulting in inadequate instrument monitoring. During this period the aircraft's speed diminished from 80 to 35 kts and the descent rate increased rapidly.The F/O was first to recognise the low speed sbut too late to avoid the crash. 

If the crew had been using a PiCMA procedure, it is likely that

1) there would have been a more detailed discussion of the auto-flight approach procedures 

2) the Captain would have had more opportunity to consider the weather reports and deteriorating conditions 

3) the Captain would have had more opportunity to consider alternatives including possible diversion

4) there would have been uninterrupted monitoring of the flight instruments by the F/O as Pilot Flying during the instrument approach phase, and a greater likelihood that the aircraft would be levelled at the MDA.   

5) There would have been more effective monitoring by the PM during the instrument approach as the cockpit authority gradient would have been reversed. 

6) It would have been harder for the Captain's plan continuation bias to dominate events as he would not have been in physical control as the aircraft reached the MDA.

Type: 
Super Puma
Where: 
Sumburgh, UK
Expected weather: 
Instrument
Pilot in charge: 
Capt
Early transition: 
Yes
Go-around : 
No
Damage: 
Serious
PicMA potential: 
Major
Year: 
2013
Time: 
Day
Deterioration: 
Yes
Vert Guidance: 
None
Both Head Up: 
Yes
LoC: 
Yes
Fully prepared: 
No
Actual Weather: 
Low Cloud
Autopilot : 
Y
CCAG: 
High