Raiman O. Diagnosis as a Trial: The Alibi Principle. IBM Scientific Center (1989).  Home/Search   Document Not in Database   Summary   Related Articles   Check

....of M= # ab .Then fM jabjjM 2M 0 g is the set of all minimal diagnoses of #SD;OBJ;OBS#. The proof of Proposition 5 can be found in [9] While previous results on the relation of diagnosis and F. Steimann et al. Model Based Diagnosis for Open Systems Fault Management 9 circumscription [1, 21] focus on formalizing stronger forms of explanation (than provided by consistency based diagnosis) this proposition establishes an interesting connection between consistency based diagnosis and circumscription, because M is the set of all models obtained by circumscribing SD#OBS in ab, while ....

Olivier Raiman. Diagnosis as trial - the alibi principle. In International Model-Based Diagnosis Workshop, Paris, July 1989.

....ways in which a component can fail. In general, a component may have several failure states. An abnormal or gate, for example, might have the output stuck at 1 (s1) or stuck at 0 (s0) org( s1, 0, 0, 1 ) org( s0, 0, 1, 0 ) org( s0, 1, 0, 0 ) org( s0, 1, 1, 0 ) The exoneration principle [17] is a special case of the strong fault model. It specifies as abnormal any behavior different from normal: org( ab, X, Y, Z ) or( X, Y, Z ) Next we define the concepts of a diagnosis and a conflict for hSD, COMPS, OBSi, assuming that an observation, a ground instance of OBS, is given. In ....

Raiman, O. Diagnosis as a trial: the alibi principle. Report, IBM Scientific Center, Paris, 1989.

....fail. In general, a component may have several failure states. An abnormal or gate, for example, might have the output stuck at 1 (s1) or stuck at 0 (s0) org( ok, X, Y, Z ) or( X, Y, Z ) org( s1, 0, 0, 1 ) org( s0, 0, 1, 0 ) org( s0, 1, 0, 0 ) org( s0, 1, 1, 0 ) The exoneration principle [Raiman, 1989] is a special case of a strong fault model. It specifies as abnormal any behavior different than normal: org( ab, X, Y, Z ) or( X, Y, Z ) In the case of combinatorial circuits, one can formulate different models in terms of boolean expressions in CLP(B) Mozeti c and Holzbaur, 1991b] ....

Raiman, O. Diagnosis as a trial: the alibi principle. Report, IBM Scientific Center, Paris, 1989.

....0 0 fail impossible 0 1 1 ab 1 0 0 ok 1 1 W unknown Table 1: A decision table encoding the state S of a component as a function of normal (F ok ) and abnormal (F ab ) behavior. F = 1 denotes that the behavior is achieved, while F = 0 denotes that the behavior is not achieved. An exoneration model (Raiman 1989) is a special case of a strong fault model and specifies as abnormal any behavior different than normal: inv( ok, 0, 1 ) inv( ok, 1, 0 ) inv( ab, X, X ) A strong fault model (Struss Dressler 1989) specifies all the possible ways in which a component can fail. In general, a component may ....

Raiman, O. (1989). Diagnosis as a trial: the alibi principle. IBM Scientific Center, Paris.

....and relies on the assumptions discussed in the previous section. Abduction [5, 2] is used to search for explanations of abnormal signals: if a signal s is abnormal, then at least one of the components or signals on which it depends must be abnormal. Corroboration (see the alibi principle in [7]) is used to exonerate components: if a signal s is normal, then all the signals and components on which it depends are assumed to be normal (notice that in this way idea cannot deal properly with masking faults, which, anyway, are not very common in the domain of application of the system) ....

O. Raiman. Diagnosis as a trial: The alibi principle. In Proc. Int. Workshop on Model-based Diagnosis, Paris, 1988.

....can fail [Struss and Dressler, 1989] In general, a component can have several failure states. An abnormal or gate, for example, might have the output stuck at 1 (s1) or stuck at 0 (s0) org( s1, 0, 0, 1 ) org( s0, 0, 1, 0 ) org( s0, 1, 0, 0 ) org( s0, 1, 1, 0 ) The exoneration principle [Raiman, 1989] is a special case of the strong fault model. It specifies as abnormal any behavior different from normal: org( ab, X, Y, Z ) or( X, Y, Z ) Next we define the concepts of a diagnosis and a conflict for hSD, COMPS, OBSi, assuming that an observation, a ground instance of OBS, is given. In the ....

Raiman, O. Diagnosis as a trial: the alibi principle. Report, IBM Scientific Center, Paris, 1989.

....in the paper that if a system description is given in the form of a logic program, then the fault notions of consistency based diagnosis and declarative debugging coincide. 7.6. Trial based software diagnosis The work of Hunt [16] transplants Raiman s trial approach for model based diagnosis [27] to the software domain. The approach is likened to the incrimination and exoneration of suspects in a justice trial. The system generates candidate diagnoses, with search being focused by removing suspects which have an alibi from the candidate list. The concept is to generate the system ....

Olivier Raiman. Diagnosis as trial - the alibi principle. In International Model-Based Diagnosis Workshop, Paris, July 1989.

No context found.

Raiman O. Diagnosis as a Trial: The Alibi Principle. IBM Scientific Center (1989).

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC