A novel approach in fuzzy bowtie analysis applying Takagi–Sugeno inference for risk assessment in chemical industry

The Bowtie technique is a powerful tool in risk analysis. From the semiquantitative perspective, logic gates might be used in system modelling to calculate the consequence frequencies and risks. However, the model results are affected by the uncertainty that can emerge from subjective evaluation, data variation, or lack of data. Fuzzy logic helps to address system uncertainty and intuitively apply specialist knowledge. Many investigations related to fuzzy fault trees, only use fuzzy sets and fuzzy mathematical relations to lead with uncertainty and still obtain equations that resemble those that originated from the traditional quantitative method. This work proposes a new approach using fuzzy sets and a rule-based inference system, aiming to deal with uncertainty. It also allows for the insertion of specialist knowledge about the system. A comparison with another fuzzy bowtie study (not rule-based) is performed to validate the model and also to discuss the benefits of the proposed modelling.