Italiano (Italian) English (Inglese)
domenica, 17 dicembre 2017


Dettagli Pubblicazione
Autori:Daniele Codetta Raiteri
Luigi Portinale
Area Scientifica:Diagnosis
Uncertain Reasoning
Probabilistic Graphical Models
Dependability and Reliability
Titolo:A Petri Net-Based Tool for the Analysis of Generalized Continuous Time Bayesian Networks
Apparso su:Theory and Application of Multi-Formalism Modeling
Tipo Pubblicazione:Chapter of Book
Sommario:A software tool for the analysis of Generalized Continuous Time Bayesian Networks (GCTBN) is presented. GCTGBN extend CTBN introducing in addition to continuous time-delayed variables, non-delayed or “immediate” variables. The tool is based on the conversion of a GCTBN model into a Generalized Stochastic Petri Net (GSPN), which is an actual mean to perform the inference (analysis) of the GCTBN. Both the inference tasks (prediction and smoothing) can be performed in this way. The architecture and the methodologies of the tool are presented. In particular, the conversion rules from GCTBN to GSPN are described, and the inference algorithms exploiting GSPN transient analysis are presented. A running example supports their description: a case study is modelled as a GCTBN and analyzed by means of the tool. The results are verified by modelling and analyzing the system as a Dynamic Bayesian Network, another form of Bayesian Network, assuming discrete time.