Development of cause-effect dependence model of undesirable events using Bayes network

In the paper the method of cause and effect analysis of undesirable events using the Bayesian networks is presented. For the analysis, due to the complexity of the calculations, it is proposed to use Java Bayes program as a free and simple tool to support Bayesian analysis. Bayesian estimation allows to identify the probability of the event occurrence. For this reason its use was proposed to determine the safety probability. Using Bayes' theorem is also possible to modify initial judgement about the situation with the use of a priori probability so that a new situation described by a posteriori probability arises. In this sense, by Bayes' theorem the data can be sequentially processed, including considerations for newer information, and thereby create a more reliable basis for decision making for the system operator. In the paper, the methodology was presented, which can be extended in order to improve the detection and monitoring of undesirable events in infrastructure.

---

Opracowanie ze środków MNiSW w ramach umowy 812/P-DUN/2016 na działalność upowszechniającą naukę (zadania 2017).