Quantitative Study over the Probability of HAZOP Analysing Technique in Petro-Chemical Plants

Jian-min FU,Dong-feng ZHAO,Guo-ming CHEN,Le-ping ZHOU,Bin ZHANG

DOI: https://doi.org/10.3969/j.issn.1009-6094.2008.06.032

2008-01-01

Abstract:The present paper is intended to present our Hazard and Operability (HAZOP) analysis method and analysis result in terms of systematic identification of conceivable deviation from the point of view of a designer. Generally speaking,HAZOP is taken as a qualitative method mixed. As is known,it is of great help to quantify the deviation created by way of systematic HAZOP analysis for risk control. Comprehensively studying the characteristics of the HAZOP technique used in petro-chemical industry,the paper has done a detailed discussion over the feasibility of the quantitative method. Seeing the fact that there has not been available accidents and deviations statistics data in the country and the complexity of the operation conditions of petro-chemical plants,the available methods for carrying out the quantitative analysis of HAZOP at home are to be put into practice along with the audit of the expert advice by transforming the deviations resulted from HAZOP analysis into the factors of probability. For this purpose,we think it convenient to present a flow chart of quantification of the probability of HAZOP deviation to a group of experts from different specialties as a think-tank so as for them to fully display their talents and specific knowledge,experience,judging ability and working attitudes. In this way,HAZOP method has been introduced to confirm the influence weight of experts from different specialties. And then AHP is analyzed according to the experts' subjective assessments,the probability of deviation in HAZOP experts' natural languages turned into the corresponding fuzzy numbers. And consequently,the fuzzy numbers located at [0,1] interval would be transformed into fuzzy failure probability by left-right fuzzy sequence. The results of our analysis indicate that the deviation probabilities of equipment failure prove to be very close to the existent failure date. Moreover,the proposed method is true to reflect the great quantity of actual probability such as human error out of control in the analysis process,if it is difficult to calculate. In addition,it proves possible to work out the deviation probabilities in HAZOP simultaneously in the computer with no additional time needed highly efficiently.