TY - JOUR
T1 - pyBNBowTie: Python library for Bow-Tie Analysis based on Bayesian Networks
AU - Zurheide, Frank T.
AU - Hermann, Eckehard
AU - Lampesberger, Harald
N1 - Conference code: 2
PY - 2021
Y1 - 2021
N2 - In addition to more conventional but often less precise methods, the risk assessment as part of the risk management process can be performed with the bow-tie analysis method. A bow-tie analysis describes the effects of causes on a top event and the resulting consequences. Bayesian networks, on the other hand, offer a mathematically concise way of describing dependencies between events under uncertainty. The mapping of bow-tie analysis into Bayesian networks is intended to make their superior calculation options available. While the mapping algorithm of a bow-tie method into a Bayesian network is described in the literature, no computer program carrying out this mapping has been found so far. In this text, a Python library, that is validated using published examples, is presented and made publicly available for mapping bow-tie methods into Bayesian networks.
AB - In addition to more conventional but often less precise methods, the risk assessment as part of the risk management process can be performed with the bow-tie analysis method. A bow-tie analysis describes the effects of causes on a top event and the resulting consequences. Bayesian networks, on the other hand, offer a mathematically concise way of describing dependencies between events under uncertainty. The mapping of bow-tie analysis into Bayesian networks is intended to make their superior calculation options available. While the mapping algorithm of a bow-tie method into a Bayesian network is described in the literature, no computer program carrying out this mapping has been found so far. In this text, a Python library, that is validated using published examples, is presented and made publicly available for mapping bow-tie methods into Bayesian networks.
KW - Bayesian network
KW - bow-tie analysis
KW - risk analysis
KW - risk assessment
UR - http://www.scopus.com/inward/record.url?scp=85101748487&partnerID=8YFLogxK
U2 - https://doi.org/10.1016/j.procs.2021.01.172
DO - https://doi.org/10.1016/j.procs.2021.01.172
M3 - Conference article
SN - 1877-0509
VL - 180
SP - 344
EP - 351
JO - Procedia Computer Science
JF - Procedia Computer Science
IS - 180
T2 - International Conference on Industry 4.0 and Smart Manufacturing
Y2 - 23 November 2020 through 25 November 2020
ER -