A runway incursion occurs when an aircraft is on a runway when it is not supposed to be there. Runway incursions are a decades-old problem. The runway incursion between two Boeing 747s at Tenerife airport in 1977 is still the worst accident in aviation history. Despite efforts to mitigate runway incursions, the number of incursions continues to grow. According to FAA data, today, on average, five incursions occur each day in the United States. Fortunately, most of these runway incursions were near-misses or incidents that did not result in injuries or aircraft damage. However, we cannot count on fortune to prevent another deadly accident.
While the COVID-19 crisis has slowed air traffic, the industry is optimistic about recovery and return to the growth in air traffic we have seen over the past decade. With this growth comes the potential for more runway incursions. Therefore, we must develop better ways of preventing incursions. Runway incursion incidents are one way to learn more about how we can prevent similar incidents in the future and reduce the probability of serious accidents. Unfortunately, most incident reports lack detailed information on the causes of runway incursions. In the United States, trained investigators at the National Transportation Safety Board investigate aviation accidents, but not most incidents, including incursions. Incursion incidents are reported by the air traffic controllers on duty at the time of incursion to the FAA. While most controller reports explain what happened, they often do not explain why the incident happened. We need deeper insight into why incidents occur so that we can develop more effective measures to reduce incursions.
Expert systems, which provide tailored questions and guidance to users, have proven useful in other fields like medicine. We are designing an experiment to investigate whether they can also help air traffic controllers provide more useful information in their reports, despite their lack of accident investigation training. For this experiment, we are developing an alternate reporting tool that provides tailored questions that aim to guide and encourage controllers to look deeper into incursions. The experimental tool guides controllers into answering three major questions: what happened (which aircraft were where, and when), how it happened (e.g., controller gave the wrong instruction), and why it happened (e.g., controller was fatigued). We will conduct think-aloud protocols and observe how controllers interact with the current and alternative reporting forms and record their report inputs and opinions.
Our findings will help explain what aids, and what detracts from, effective incursion investigation and reporting. This research may help the FAA and airports improve controller training as well as suggest ways of helping controllers prepare more useful reports. Such reports can help the aviation community identify the cause of human errors leading to incursions, and develop more effective mitigation strategies, ultimately saving lives.