*Special session on Bayesian Networks in Dependability *
IFAC Workshop on DCDS (Dependable Control of Discrete Systems
http://dcds09.poliba.it/DCDS09/Home.html), Bari, June 10-12, 2009
*_Session co-chairs: _
Stefania Montani
*Dipartimento di Informatica*
*Universita' del Piemonte Orientale
Via Bellini 25/g
15100 Alessandria, Italy
email: [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>
* Philippe Weber*
Université Henri Poincaré, Nancy
Rue Jean Lamour
54519 Vandoeuvre-lès-Nancy
email: [EMAIL PROTECTED]
_*Call for papers*_
Bayesian Networks (BN) have been proposed in the field of Artificial
Intelligence as a probabilistic framework for reasoning under uncertain
knowledge. The robustness and flexibility of this formalism are
demonstrated by the wide spectrum of real world problems in which they
have been successfully tested.
The area of dependability has to be included in the range of BN
application domains, since, in recent years, some works have started to
investigate the capabilities of this methodology in modeling and
analyzing complex and safety critical systems. In particular, with
respect to traditional dependability analysis formalisms, BN show an
increased modeling power: they allow to deal with multi-valued
variables, local dependencies among components, and noisy interaction
among component behavior in a natural way -- three issues that can not
be easily accounted for by Fault Trees, for instance. In addition, a
general inference mechanism (combining prediction as well as diagnosis)
can be naturally performed on them, even if evidence is gathered during
analysis.
Several extensions of the BN formalism also appear to be of great
interest: among them Dynamic BN and temporal BN in general for dealing
with dynamic system behavior, and object-based representations for
dealing with parametric representations. BN and their extensions also
appear to be fruitfully applicable in the area of control and automation.
The goals of this session will be to collect present approaches to
dependability applications of BN, and show case studies and practical
experiences.
_*Topics of interest*_ will include (but are not limited to):
- BN for modeling and analyzing safety critical systems
- BN and BN extensions for modeling and analyzing systems showing
complex dependencies
- BN and BN extensions for process monitoring
- comparisons among the application of BN and other more classical
methodologies
- comparisons among different BN approaches to dependability problems
- case studies in using BN for dependability analysis
- case studies in using BN in control and automation
*_Submissions:_*
The paper submission will take place through the DCDS website, according
to the conference deadlines (
<http://dcds09.poliba.it/DCDS09/Importantdates.html>http://dcds09.poliba.it/DCDS09/Importantdates.html)
<http://dcds09.poliba.it/DCDS09/Importantdates.html>
--
Stefania Montani, Ph.D.
Assistant Professor
Dipartimento di Informatica
Universita' del Piemonte Orientale
Via Bellini 25/g
15100 Alessandria, Italy
phone: +39-0131-360158
fax: +39-0131-360198
email: [EMAIL PROTECTED]
http://www.mfn.unipmn.it/~stefania
_______________________________________________
uai mailing list
uai@ENGR.ORST.EDU
https://secure.engr.oregonstate.edu/mailman/listinfo/uai
