Introduction: Risk assessment of hazardous processes is the priority of risk management. Layer of protection analysis (LOPA) is one of the most popular methods used for risk assessment. Due to the insufficient information or uncertainty in failure rates (PFD) of protective layers, risk assessment based on the conventional LOPA can result in error in calculations. In this study, we tried to use the fuzzy set theory to evaluate the ammonia receiving tank safety, using the LOPA method.

Methods: Initially, the fuzzy failure rate of protective layers were calculated using the subjective opinions of professionals. Then, by applying the fuzzy operators, fuzzy possibilities transformed to fuzzy probabilities and subsequently they were deffuzified to crisp failure rate. Afterwards, using the severity fuzzy logic, severity of the outcome event was calculated in the fuzzy form, and subsequently, fuzzy risk index was calculated using the fuzzy matrix.

Results: In the ammonia release scenario, calculated severity, probability and risk levels were determined as P: Low, S: High, and R: TNA, and PF = -2.66, SF = 3.99, RF = 3.79 (0.2 TNA, 0.8 NA) for classic and fuzzy LOPA methods, respectively. In addition, after inserting additional layers of protection, the fuzzy risk index reduced from 3.79 (0.2 TNA, 0.8 NA) to 1.92 (0.1 A, 0.8 TA, 0.1 TNA).

Conclusions: In the condition of uncertainty and lack of information relating to probability and severity of risk scenarios, the experts’ opinions can be used in forms of linguistic variables and fuzzy relations to reduce calculation errors in risk assessment as much as possible