Quantitative Analysis of Sticking Risk Based on Fuzzy Bayesian Networks

Quantitative assessment was carried out on the sticking risk in the drilling and completion stage to fill the gap in this field， and provide scientific basis and technical support for the risk control of sticking in drilling operations. Based on the analysis of causes， inducing factors and common types of sticking， a fault tree model was constructed and mapped to a Bayesian network model. Using expert knowledge and real data from an oil field in Southwest Sichuan Basin， fuzzy analytic hierarchy process （FAHP） was used to quantify the risk of sticking， and the main risk factors affecting sticking were identified by sensitivity analysis. The research results show that human error is the main cause of sticking， with its contribution rate significantly exceeding other factors. Sensitivity analysis identifies key causal factors such as improper drilling parameters， inadequate connections and long rig downtime. The constructed model can dynamically update the risk probability， thereby effectively supporting risk warning and decision-making. The research conclusions not only provide theoretical support for the oil and gas industry， but also provide practical guidance for risk management in drilling operations， and have important engineering application value.