Application of a Mistake Severity Criterion in Expert Systems Designed to Train Flight Safety Inspectors

Abstract

This article was prepared within project No. 1908-00028 "Development of a concept for building the architecture of, and selecting algorithms for, an expert system aimed to improve training efficiency for flight safety inspectors" supported with a grant awarded by the Russian Foundation for Basic Research (RFBR).The article deals with an approach to selecting and substantiating parameters characterizing the competency profile of a flight safety inspector. A brand new "error severity criterion" is proposed to assess the severity of probabilities of an inspector’s mistakes depending on possible circumstances and operating conditions. It offers an example of how such criterion is assessed by analyzing identified inconsistencies and violations in implementing monitoring and supervision procedures and assessing potential consequences that may result from an inspector’s erroneous conclusions and findings.