The investigation report of the October 2001 runway collision between a taxiing Citation CJ2 and a Scandinavian Airlines MD-87 taking off at the Linate Airport at Milan, Italy, is quite revealing. It states that although the airport surface surveillance radar was not operating at the time, coupled with slack air traffic communications procedures and inadequate airport signage, “the accident ultimately resulted from the Citation crew’s human error in taking an incorrect taxiway and straying onto the active runway. The Citation taxied across several markers, including a stop line and an illuminated bar of red stop lights, without communicating with air traffic control.”

Visibility when the Citation landed an hour earlier was estimated as being between 50 and 100 meters (164 and 328 feet). The crash killed all 114 people on both airplanes, as well as four ground workers. Like all accidents, this one shouldn’t have happened, even under the difficulties the pilot faced.

Paradoxically, runway incursions and other ground collisions–when aircraft are moving at their slowest–have been at the top of the NTSB’s “Most Wanted List of Transportation Safety Improvements” since the list’s inception in 1990, a fact underlined in 1999, when then FAA Administrator Jane Garvey said, “Taxiing on the airport surface is the most hazardous phase of flight.” And while the FAA, NASA and private industry groups had been working the problem for many years before that, runway incursions still happen, although their numbers are dropping, albeit slowly.

And what exactly is a runway incursion? The FAA defines it as “any occurrence in the airport runway environment involving an aircraft, vehicle, person or object on the ground that creates a collision hazard or results in a loss of required separation with an aircraft taking off, intending to take off, landing or intending to land.”

The FAA’s recent Runway Safety Report, a statistics-packed 86-page document put out by the agency’s office of runway safety, shows that between 1999 and 2002, total incursions reported by towered airports went from 329 in 1999, to 405 in 2000, to 407 in 2001 and then dropped to 339 in 2002. Preliminary data for last year indicates a further slight drop to 323. These numbers were further broken down in three ways.

First, incursions were classified by their severity, with category A requiring extreme avoidance action to avoid a collision or resulting in a collision. In category B, there is a significant potential for a collision; in category C, there is ample time and distance to avoid a collision; and in category D, there is little or no chance of a collision but the incident meets the broad definition of an incursion. Over the aforementioned four-year period, category A incidents accounted for 5 percent of the 1,477 runway incursions reported; category B, 10 percent; category C, 33 percent; and category D, 52 percent.

The second breakdown sorted the type of error into pilot deviations, controller operational errors or vehicle/pedestrian deviations. Here, pilot deviations led, with 58 percent of the total, followed by controller operational errors at 22 percent and vehicle/pedestrian deviations at 20 percent.

The third breakdown divided incursions into those committed at the nation’s top 35 commercial airports, and those at the 35 busiest general aviation airports. Predictably, pilot deviations at general aviation airports were higher, at 63 percent, than commercial pilot deviations at commercial airports at 51 percent.

But whichever way the numbers were presented, no silver bullet appeared, and the FAA report wisely avoids pointing fingers in any specific direction. On the other hand, the inescapable but unstated conclusion is that human error accounts for virtually all incursion events. To this end, several technologies have been introduced at least to minimize the problem.

Tower Surveillance

Airport surface detection type 3 (ASDE-3) is a radar surface surveillance system that shows moving aircraft and vehicles as simple “blips,” without individual idents, on a controller’s screen in the tower. Thirty-four ASDE-3 systems are in operation, mainly at large commercial airports.

Airport movement area safety system (AMASS) is a computer program that assigns ident tags to specific targets on the ASDE-3 display and alerts controllers when it calculates a potential risk of collision. But the NTSB cautions that the alerts can provide controllers with as little as eight to 11 seconds to warn the pilots involved.
ASDE-X is a newer surface surveillance system using more accurate X-band radar, as well as “multilateration” technology that adds ident tags to all targets on the controllers’ display. ASDE-X is now installed at 25 FAA towers, and a further 25 systems are on order.

Aircraft Warning Systems

While the FAA has traditionally given tower controllers the responsibility for advising pilots of potential incursions and collisions, the NTSB has consistently called for pilots to receive direct warnings from onboard systems.

Automatic dependent surveillance-broadcast (ADS-B) provides a flight-deck display of all other ADS-B-equipped aircraft on the surface or in the local airspace. Various implementations have been developed (by UPS/Garmin, NASA and others) and evaluated.

Multilateration systems, which are being researched by NASA and Rannoch’s PathProx, detect signals from ATC and mode-S transponders, TCAS, ADS-B, military IFF and other aircraft emissions to locate proximate aircraft and determine collision threats.

Honeywell’s runway awareness and advisory system (RAAS) uses GPS and highly accurate airport maps to clearly show an aircraft’s position on approach and while taxiing. It does not, however, show other aircraft.

The surface guidance system (SGS) from Rockwell Collins’s Flight Dynamics division provides precision taxi guidance, showing runways, turn-offs, stop bars and taxiways via a HUD. The addition of an enhanced vision system would allow other aircraft and vehicles to be detected.

Enhanced vision systems (EVS) by CMC Electronics, Kollsman and Max-Viz provide detection of aircraft, vehicles and surface objects in low visibility and at night. But unless linked with an SGS or a similar system, EVS by itself does not provide airport surface taxi guidance.

Yet none of these technologies should ever let us forget our flight instructor’s classic admonition when, as student pilots, we released the brakes and taxied an aircraft for the first time: “When in doubt, stop.” It might have made all the difference in Milan.