B.H. Far's Home Page

Theme 3 (old):
Qualitative Sensitivity Analysis


We introduce the Qualitative Sensitivity Analysis (QSA) technique and its applications for systems modeled in rule based causal format. The rule set is composed of interacting atomic propositions. First, we introduce the Modified Causal Ordering (MCO) technique to derive ordering among the propositions. The result of MCO are complete subsets of propositions and an Ordinal Reachability Graph (ORG). Each complete subset is identified by its ranking code. In QSA, propositions stand for the landmarks of the qualitative variables. QSA begins with the ORG and replaces the propositions with their equivalent landmarks. Sensitivity is defined qualitatively in terms of possible shift to the neighboring landmarks. QSA can derive the sensitivity of the higher rank landmarks due to perturbation affecting the lower rank ones. Here we introduce the application of QSA for performance evaluation of discrete event dynamic systems (DEDS) and also generating diagnostic rules for concurrent malfunctions in rule based systems. We demonstrate the approach through two detailed illustrative examples.

Additional Documents

For further information on the theme see below.
PDF version of the detailed document is available.
Please click on the pdf button to view it.

B.H. Far and M. Nakamichi,
Qualitative Fault Diagnosis in Systems with Nonintermittent Concurrent Faults: A Subjective Approach,
IEEE Transactions on System, Man and Cybernetics, vol. 23, no. 1, pp. 14-30, Jan/Feb 1993.

B.H. Far,
A Research on Applications of Qualitative Reasoning Techniques in Human Acts Simulation Program,
Technical Report JAERI-M 92-058 79 pages, April 1992.

B.H. Far, M. Nakamichi and M. Sampei,
Qualitative Sensitivity Analysis,
Japanese Artificial Intelligence Journal, vol. 6, no. 1, pp. 84-95, January 1991.

B.H. Far, M. Nakamichi and M. Sampei,
Automatic Diagnostic Rule Generation for Rule-based Controllers by Qualitative Sensitivity Analysis,
Transactions of Electerical Engineering Society, vol.110.c, no. 10, pp. 661-669, October 1990.