APPLICATION OF FUZZY LOGIC IN RISK ASSESSMENT OF FAILURES AND ACCIDENTS FOR TRUNK OIL PIPELINES

Abstract

The majority of existing methods of risk assessment of trunk oil pipelines is based on probability and classical set theory. These methods do not take into consideration the fact that any complex system, such as trunk oil pipelines, is a dynamic system with a set of uncertain data. The results of risk assessment (likelihood), obtained by the traditional methods, don’t reflect the real situation about the condition of system. The article presents a methodical approach to identification of risk factors sources and assessment of the level of risk based on system analysis and fuzzy set theory, and this approach allows us to receive output conclusion about risk level in dynamic area of a complex system. The principles of fuzzy logic in the evaluation of diverse parameters of risk factors allow bringing them to a qualitative common denominator. As a result, we can perform comparative research of separate risk factors and also summarize them for assessment of current risk level of trunk oil pipelines. We reviewed causes of incidents and failures on trunk oil pipelines and have identified the most important risk factors that have greatest consequences. Risk factors in our method are determined by fuzzy set rules, which are a result of data analysis (accident reports, failure records for the similar pipelines) and expert’s evaluation of related input parameters. Risk factors and their parameters are classified in accordance with their relation to different environment and life-cycle stages. The article presents an example of evaluation of risk factor . Consequently, the indicator of risk factor was calculated in accordance with procedure of Mamdani fuzzy inference system (using the MATLAB Fuzzy Logic Designer). This method of risk assessment brings high opportunities for identification of the pipeline segments, where the risk level is most probably high and helps the decision maker to take remedial actions to mitigate potential risks. The operator can more effectively manage resources and improve efficiency of actions. Ability to reflect the current risk level is key advantage of our approach to risk assessment of risk management of trunk oil pipelines.