Intelligent Decision Support System When Assessing Risk Significance

Abstract

The paper presents an intelligent information system for solving the urgent problem for determining the significance of risk in the activities of testing laboratories. The algorithm of the system functioning is based on fuzzy logic inference of the assessment of risk significance. The expediency of this approach is stipulated by the expert nature of information about the parameters of risks, the availability of subjective judgements and fuzzy knowledge. As a result of study of the object domain, the input attributes have been established – the possibility of risk and the severity of risk consequences which determine the output parameter – the significance of the risk. Based on the expert knowledge, the corresponding linguistic variables have been introduced, basic and extended term-sets have been established, membership functions have been constructed. As a result of processing and analyzing the consistency of the expert information, the optimal models of linguistic variables have been chosen. The system of production rules being in the basis of fuzzy inference of the solution – assessment of risk significance – have been created. Recommendations for risk management have been developed for each level of the risk. On the basis of this algorithm, an original program has been created which makes it possible to instantly get information about the significance of the arising risks and take necessary measures under the conditions of continuous operation of the laboratory. The illustrations of the program running have been presented.KeywordsIntelligent information systemRisk managementFuzzy expert knowledgeLinguistic variableMembership functionIndices of consistency