Buch, Englisch, 200 Seiten, Format (B × H): 145 mm x 210 mm
Buch, Englisch, 200 Seiten, Format (B × H): 145 mm x 210 mm
ISBN: 978-3-89722-866-5
Verlag: Logos
This thesis describes model-based test generation w.r.t. repair aspects. In the first part of this thesis we present a formal theory for modeling actions and observations, which provides a foundation for model-based diagnostic reasoning. Then we describe an algorithm for test generation, i.e. for proposing setup-actions and observations that lead to the discrimination among diagnostic candidates.
The algorithm is based on decision trees and it uses probabilities to rank tests according to their information-theoretical contribution w.r.t. the repair process. In the second part of this thesis we present the design and implementation of a knowledge-based authoring system. This authoring system is a tool that allows the user to edit device models by means of a text-processing interface. The model can be used for diagnostic applications as described in the first part of the thesis. Both, the test generation algorithm and the authoring system, have been evaluated w.r.t. typical requirements from the process of creating repair manuals for passenger vehicles.
