Operator Support in Traffic Management: A Heuristics Model and Experimental Study

Abstract

The identification of problems from numeric traffic measurements is an important part of control center activities in ATMS (Advanced Traffic Management Systems). However, an information modeling process that relies solely upon ‘traditional’ quantitative data analysis does not reflect faithfully the actual methods used by human operators. In addition to common-sense knowledge and specific contextual information, operators also use various heuristics and rules-of-thumb to supplement the numerical analysis. This paper describes an experiment to examine the effectiveness of an expert system that integrates quantitative and qualitative traffic information using a human-centered knowledge system design. The system's performance was investigated using a data suite of real traffic scenarios; the statistically significant results showed that the integrated process had superior performance compared to the ‘traditional’ quantitative analysis running alone.