Introduction: Due to the presence of extreme hazard sources and high intrinsic risk in refineries and process industry sectors, different layers of protection are being used to reduce the risk and avoid the hazardous events. Determining Safety Integrity Levels (SILs) in Safety Instrumented Systems (SISs) helps to ensure the safety of the whole process. Risk Graph is one of the most popular and cost effective techniques to do so. Despite Risk Graph simplicity it’s a qualitative method which is highly subjective and suffers from interpretation problems that can lead to inconsistent or conservative SILs.

.

Material and Method: In this paper, Improved Risk Graph (IRG) method was presented and evaluated, and using Fuzzy Logic a novel approach namely Fuzzy Improved Risk Graph (FIRG) was suggested. In the proposed method consequence levels which were defined as qualitative terms were transformed into quantitative intervals. Having those numerical values, risk graph table was converted to a quantitative one. Finally, applying the presented approach and using three experts’ opinions and attributing weight factors, an ultimate numeric value was produced.

.

Conclusion: as a result of establishing the presented method, identical levels in conventional risk graph table are replaced with different sublevels that not only increases the accuracy in determining the SIL, but also elucidates the effective factor in improving the safety level and consequently saves time and cost significantly. The proposed technique has been employed to develop the SIL of Tehran Refinery ISOMAX Center. IRG and FIRG results have been compared to clarify the efficacy and importance of the proposed method