reducing "Crew-caused"
approach and landing
accidents 

Pilot-in-charge Monitored Approach

Individual pilot "quality".

Since human pilot error in the flight deck is still the predominant immediate cause of airline accidents, the approach to dealing with it has typically been to provide more crew training and education intended to improve crew performance. Even though it may have an emphasis on crew interaction, it is still essentially about improving the "quality" of individual crew members so that they interact better. 

But is this sufficient? Can we just rely on training to deal with the type of tactical errors/errors of omission that the NTSB noted as being particularly difficult for the PM to correct?   

How well are we actually doing with CRM?

One 2000 academic study noted that  "unfortunately, the successful introduction of CRM training into many airlines has not been mirrored by any noticeable reduction in the number of incidents and accidents that stem from crew coordination and communication problems."  A 2002 study also concluded that there was little formal evaluation of CRM training effectiveness. Those studies were over a decade ago - but in 2013 a leading manufacturer's chief pilot still posed the question "Are we doing as well with CRM as we think we are - or are we paying lip service?"

Most countries now require CRM training at least for air transport operations, to meet the intentions of accident investigators recommendations and similarly motivated state regulatory organisations.  Regulations might say that for example operators should ensure that "all pilots have training in team management, communications, situation awareness, decision-making, and recognition of the resources available to assist the crew in the safe, efficient completion of any flight operations activities".

Unfortunately, what is actually done to achieve the necessary behaviour change is often simply unrealistic. An airline's approved initial CRM training may consist of a 6-hour CRM module with a PowerPoint presentation on topics including "the captain's authority, team building, decision behavior, inquiry and assertion, conflict resolution, workload management, and situational awareness", and  a couple of hours of videos, all self-taught with no instructor guide.

CRM Training: Reality v Theory 

Accident investigations continue to show a startling discrepancy between good intentions and actual practice. In one recent Asian accident to an airline that met the wording quoted above, the Captain had passed a recent proficiency check, although "graded at minimum standard at CRM/Threat & Error management and  workload management". The F/O had met minimum standards and was also assessed "satisfactory". Nevertheless the F/O was unable to provide effective monitoring due to the excessive CCAG.

In another recent Canadian accident report, it emerged that the basic CRM training actually given to the F/O was limited to a total of 3 hours 10 min of potential tuition time in a classroom.  When a similar course was observed, 5 of 8 mandatory topics were omitted. In fact this F/O's basic CRM training was essentially one half of a day which also featured briefings on the company's Safety Management System, and on various of its security requirements.  Again, the F/O was unable to provide effective monitoring due to the excessive CCAG.

A 2011 European accident report showed that on the Captain's recent command training, 75% of the CRM element had been omitted.

A 2015 report into a US executive jet accident in which 7 passengers died showed many organsational defects, including CRM training consisting of a PowerPoint showing of headings from the appendix to the FAA circular on the subject, followed by a 10 question test with a passing grade of 80%. The Captain had scored 40% and the F/O 70%, but the Captain's was marked as 100%. The report noted that "Areas of CRM in which the crewmembers were deficient according to their CRM test results included responsibilities of the PIC, flight deck management, and aeronautical decision-making. Significantly, the captain’s errors during the flight were in these deficient areas, demonstrating that the CRM training the crew-members received was ineffective." 

It seems from many other reports that regardless of where an operator is based, in practice little reliance can be placed on the actual effectiveness of regulatory obligations to provide CRM training.