Authors: Roeland Scheepens, Steffen Michels, Huub van de Wetering, Jarke van Wijk
Abstract: We present a method to visually explain the rationale of a reasoning engine that raises an alarm if a certain situation is reached. As a case study we look at the maritime safety and security domain. Based on evidence and a reasoning structure, the engine concludes with a certain probability that, e.g., the vessel is an environmental hazard. This engine is part of a larger safety and security system manned by an operator who makes decisions based on the output of the reasoning engine. To support decision making we visualize the rationale, an abstraction of the reasoning structure that allows users to understand why the conclusion has been reached, and display the evidence in a color-coded matrix that easily reveals if and where observations contradict.