When the first Flight Management Systems (FMS) were being introduced many years ago, a common phrase heard in the cockpit was, "What's it doing now?" Up to that point it was the pilot who analyzed the raw data and made all the decisions about when to change over to a new fix or a new navigation source. While it could sometimes be a challenge to put together all the available information to figure out where you were or where you were supposed to be going, it was up to you. In a crew situation the pilot not flying might occasionally wonder "What's he doing now?" about the pilot flying the airplane, but he could then clarify the situation by asking the pilot.
The introduction of the FMS was like adding a person to the cockpit. Once programmed by the pilot, the FMS would make decisions such as when to change to the next fix all by itself. Unlike a person, you couldn't ask it questions about what it was doing or why it was doing it. This led many pilots to feel like there was now another "brain" in the cockpit-a somewhat mysterious, even malicious creature that had taken over their airplane as they sat there wondering "What's it doing now?"
Flight Management Systems are now common in most airliners, and even two-seat trainers are being equipped with GPS navigation systems that are close to being an FMS in their capabilities. An accident in 2004 provides a chilling example of how easy it is to lose track of what an FMS or GPS is doing and blindly follow the guidance it is providing. A Beech King Air 200 was approaching Martinsville/Blue Ridge Airport (MTV) in Martinsville, Virginia. The name "Blue Ridge" immediately suggests higher terrain surrounding the airport, and a quick look at the approach plate shows an airport elevation of 942 feet with higher terrain (up to 5,100 feet) within 25 nm in all directions. The NTSB report states that "as the airplane approached MTV, an air traffic controller advised the flight crew that the airplane was second in line for the localizer 30 approach." The crew was instructed to hold as published on the localizer course at 4,000 feet MSL with an expected 28-minute delay in the holding pattern. At this point it is probably safe to assume that the crew was at least a little frustrated that they would have to hold for almost half an hour after a less than 30-minute flight. They requested and received approval for five-mile legs and got ready for the hold.
A few minutes later the controller suddenly cleared them for the approach. The NTSB report states that they "completed a continuous right turn toward the inbound course and crossed BALES LOM at an altitude of 3,900 feet." There was one small problem. The altitude specified on the approach plate for crossing BALES is 2,600 feet, so they were 1,300 feet above the specified altitude with only five miles to the missed approach point (MAP) in which to lose a total of 2,380 feet and still arrive at the end of the runway in a stable configuration from which they could make a safe landing.
Instead of this "slam-dunk" approach, the airplane gradually descended until it reached 2,600 feet as it approached the MAP. At that point, instead of executing the missed approach, the airplane continued to descend to 1,400 feet, which was just slightly higher than the minimums specified for a straight-in approach using DME. The NTSB report says they continued about eight miles past the airport and then initiated a climb straight ahead. They had climbed about 700 feet over the next two miles when they declared a missed approach on the approach control frequency. Shortly after the controller acknowledged the call, the airplane struck mountainous terrain at an elevation of about 2,400 feet.
How could a professional flight crew get so disoriented that they would fly eight miles past the airport and then continue straight ahead rather than initiating the published missed approach procedure, which called for a climbing right turn to 2,600 feet back to BALES? As the NTSB investigators analyzed the data and tried to figure out what might have caused this accident, they noticed that every action the crew took seemed to be displaced by about 5 nm, which is exactly the distance from BALES to the MAP located at the end of the runway. It seemed as if when they reached the MAP at the airport, they actually thought they were at BALES.
Interviews with other company pilots provided clues that suggested a likely scenario. The airplane was equipped with an IFR-capable GPS, however it was not actually certified for IFR operations and the database was not current. The pilots said they would typically program the LOM and the airport as waypoints. The GPS would autosequence to the airport as the airplane passed the LOM, providing distance and navigation information to the airport. If the airplane crossed the LOM twice, such as in this case, the GPS would autosequence to the airport on the initial crossing.
Besides the obvious failure to properly execute the published instrument approach and missed approach procedures, the NTSB listed as a contributing cause "the flight crew's failure to use all available navigational aids to confirm and monitor the airplane's position during the approach." It seems likely that with an expected hold of almost half an hour, the crew had not mentally prepared for the approach yet. When the approach clearance was unexpectedly received a few minutes later, the best course of action would have been to proceed outbound in the holding pattern while they got ready for the approach and started a descent, and then turn back towards the airport only when they had enough distance to arrive at the specified altitude before reaching BALES. Instead they immediately initiated the approach before they were mentally prepared or had set up the cockpit. In the rush to get down and get ready, they somehow missed the fact that the GPS had autosequenced and they were reading the distance from the airport rather than from BALES.
