A Belief Rule-Based Generic Risk Assessment Framework

The present work aims to propose a generic risk assessment framework for modeling, analyzing and synthesizing risk-related information with various uncertainties. Such information may be of very different nature: it can be quantitative or qualitative, and uncertain, incomplete, imprecise, conflicting, for which the traditional quantitative approach (e.g., statistical approach) does not give an adequate answer. If–Then rules place the key role in this framework which is then combined with Dempster-Shafer theory of evidence and decision theory in order to form a generic If–Then belief rule based risk assessment framework with the aim to address different types of information. The general risk assessment specification for the risk assessment of a hazard event is reviewed firstly. The belief-rule based risk assessment framework is then illustrated and outlined including rule-base representation, inference procedure, rule-base generation and multi-source synthesis aspects. This is then followed by the illustrative analysis of hierarchical structure nature of the real world risk assessment model. How can the above discussed risk assessment framework be incorporated to form a multi-layer risk assessment model is finally specified. In the proposed generic framework, various types of information from different sources can be transformed and used in the inference process. It provides a flexible and effective way to represent and a rigorous procedure to deal with hybrid uncertain assessment information to arrive at rational conclusions.