A Bayesian Belief Net Model to Evaluating Organizational Safety Risks

A Bayesian Belief Network (BBN) is a valuable tool to represent the causal relationships that exist in a given set of variables. This paper presents a methodology for organizational risk analysis for safety management. Learning a BBN from data is a difficult and resource-consuming task, we presents the implementation of a greedy algorithm that automatically constructs a BBN from a dataset of cases obtained. The resulting BBN reflect installation specific factors respect to organizational factors and show the dependencies that exist among key variables that are associated to the trip generation process.