CW247's original question.
CW247 asks a nice simple question – how does a “monitored approach” improve things?
First, what is meant by “monitored approach”? It shouldn’t need saying that there is “monitoring” in every approach procedure for a two-pilot aircraft. I define it as an SOP in which the “pilot in charge” carries out the “pilot flying” tasks for take-off and landing, but the "pilot monitoring/PNF" tasks for the approach and for normal go-arounds. I’ll call it PiCMA for short – “Pilot-in-Charge” Monitored Approach. It has no necessary limitations regarding pilot rank (captain / first officer) or association with specific weather conditions or levels of automation. While many operators implement PicMA with a variety of such criteria, in my opinion every limitation means the loss of some potential safety benefits.
These benefits are reductions in the risks from the three primary reasons for “crew-caused” approach and landing events in which serviceable aircraft are damaged or destroyed. These are
1) Premature transition from instrument flight to external visual cues, often resulting in the aircraft continuing below Decision Height without adequate guidance.
2) Poor or inadequate resource management of the approach and go-around, leading to excessive pilot workload
3) Ineffective cross-cockpit monitoring, where the pilot monitoring is unable to prevent the pilot flying from continuing an unsafe flight path.
Each of these primary reasons has several contributory elements of its own, such as plan continuation bias in (1), task saturation for (2), and authority gradient for (3). But nearly all approach and landing accidents arising from crew actions or failures (popularly known as “pilot error” accidents) contain at least one of these factors and many contain all three, sometimes reinforcing each other to create a catastrophic situation. If these are recognised in the accident report there are usually anodyne recommendations that boil down to “do what you’re told by regulation and training”.
However it’s a fact of life that there are and will always be deficiencies in regulations and pilot training. Even if these seem adequate on paper, real-world implementation is often seriously deficient. These are holes in the proverbial “swiss-cheese slices” which are underlying accident causes, but are seldom publicised as more than minor contributory factors, while the crew’s ultimate responsibility will be made very evident in any investigation report. If the pilot-in-charge delegates short-term flight path control responsibility for the most challenging part of the flight, protection against these hazards is hugely increased, and this is the fundamental point of routinely using the PiCMA procedure. You assume the worst – that everything is out to get you – until it’s proved otherwise.
For those that want data, I’ve analysed over 100 commercial transport events (accidents and incidents) from the ASN database between 1990 and 2015, going back to the original reports wherever possible, to see where PiCMA might have affected the crew actions. There are very few where it would not have some effect and interrupted the sequence of events that ended in an accident or serious incident. All this and much more is on my picma.info website, or you can PM me.
Some interesting snapshots: 59% were class 1 events (hull loss). 70% were Captain flying, 78% if you include those where the Captain over-rode the F/O when the F/O was PF and supposed to be “in charge”. 66% had autopilot in use, in 64% instrument vertical guidance was available, 41% were at night, in 39% the crew had anticipated poor weather, and in 11% the weather was worse than expected. In 46% the Captain was flying using the autopilot. The aircraft types were very roughly 28% Airbus, 41% Boeing, 29% other. (I’m open to having this data corrected and updated.)
There’s no doubt that PiCMA makes many pilots uncomfortable at first. It is certainly goes counter to an “I want to do it all, and do it my way” attitude which is common in the pilot community. It appears in quite a lot of comments about it – and more importantly is very obvious in many of the events analysed. By requiring the pilot-in-charge to specifically re-take control, PiCMA undoubtedly makes it easier to resist the temptation do the enjoyable and easy thing which turns out to be wrong, like declaring some lights or a runway glimpsed at a distance to actually be the destination and in the right relative position for a safe landing. It protects against the traps which result from poor training and inadequate regulation, or poor enforcement of regulations, in other areas.
The continued, even if rare, occurrence of “crew-caused” events is in my opinion a major reason why managements and authorities, desperate to avoid any share of blame for them, respond by putting more and more insistence on rigid adherence to maximum use of automation, to the detriment of airmanship and necessary actual aviation skills. If the pilot community wants to reverse that trend, the answer is in your own hands. There’s been a better way to avoid the booby-traps that lead to such events, but up to now most don’t want to accept it.
Happy New Year to all.....