This proposal will develop an empirically-tested model to investigate factors contributing to near and actual mid-air collisions between aircraft. The model will contribute generally to an improved understanding of the use of influence diagrams to study incident sequences and sub-system failures in an wide range of low risk-high consequence events that are important to the fields of operations research, risk analysis, and transportation. The specific applications area, involving mid-air airplane collisions, will improve our understanding of the interrelationships between human errors, collision avoidance systems on airplanes, and new air-traffic control systems. These linkages, although vital to the successful operation of the air traffic control system in the U.S., have not been addresssed satisfactorily by existing models. Previous research using probabilistic models conducted by the principal investigator suggests that the results of this study will have an influence on both the theory and the practice of modeling low risk-high consequence accidents.
