Large-scale validation of a security inspection model

Security applications of human factors are still relatively few making it important to access and use as much relevant information from other domains as possible. This paper measures the validity of human factors knowledge from industrial and aviation inspection when applied to a security task: X-ray screening of carry-on bags. A model of security as a type of inspection based on the functions and tasks involved, had not been formally tested in this domain. This paper analyzes four existing data sets plus a new experiment to measure different aspects of x-ray screener performance. The data sets were from two continents and involved a total of 564 experienced screeners and 58 novices performing an off-line simulation of the carry-on bag screening task. In addition, a new experiment using 22 screeners and 24 non-screeners examined performance for model validity. In all but one comparison, the model was proven valid.