An Ordered Logit Model of Air Traffic Controllers' Conflict Risk Judgment

In this article, the authors describe a model they have developed to model air traffic controllers' judgment of conflict risk. Starting with simple scenarios, the authors build up to create a model for more complex field situations. This work is part of the Conflict Risk Evaluation based on Expert Detection (CREED) project, which seeks to provide an expert model of conflict detection, which will have as output the expected conflict risk perceived by the controller rather than a conflict/non-conflict diagnosis based on objective threshold metrics. In a large experiment, experienced controllers confronted pairs of converging aircraft in an approach environment. The controllers were asked to provide judgments as to possible loss of separation. This article analyzes the data and presents a statistical model to predict controllers' answers under the experimental conditions. Among the considerations the authors included were human-centered analysis of conflict detection and integrating doubt management into conflict detection. The construction of the conflict scenarios is described, along with the experiment's procedure. The ordered logit models for each of the variables of the explicative variable are described, as is the integrated model, which gives a more general view. While this model can predict controllers' judgments under the experiment's conditions, more work is needed to broaden the scope of scenarios and conditions that are covered.

  • Availability:
  • Authors:
    • Averty, Philippe
    • Guittet, Kevin
    • Lezaud, Pascal
  • Publication Date: 2008

Language

  • English

Media Info

  • Media Type: Print
  • Features: Figures; References; Tables;
  • Pagination: pp 101-125
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 01105573
  • Record Type: Publication
  • Source Agency: UC Berkeley Transportation Library
  • Files: BTRIS, TRIS
  • Created Date: Jul 22 2008 4:08PM