In order to explore the main factors influencing the vibration interlock shutdown of a decarbonized lean fluid pump in a LNG plant, the occurrence law of the event has been sorted out by comprehensively analyzing the operating condition so as to guide the rectification in the next step.Based on an example of vibration interlock, the grey relational analysis method is adopted to enrich data volume and the whole process operation data before and after the event is selected for analysis.The results show that temperature, vibration, instrument damage and other causes of pump interlock are intersecting, which can not only trigger interlock but also promote each other to stop the pump.The grey relational analysis method is used for mathematical modeling to make full use of current data.According to different working conditions, the master factor and sub-factor are set flexibly, and the whole data dynamic analysis is carried out for the fuzzy qualitative interlock shutdown.Induction points have been quantified, and guidance for trouble-shooting at site has been conducted effectively to solve the bottle-neck effects due to limited analyzing methods and avoid blind pump dismantle.The economic risks of buying and replacing spare parts have been reduced and the technical management level has been improved to certain extent.