基于FMEA-FBWM-RBN的中国粮食进出口海运风险评估-《集美大学学报（自然科学版）》

文章信息/Info

Title:: Maritime Risk Assessment of China’s Grain Import-Export Based on FMEA-FBWM-RBN

Author(s):: SONG Yunting1; QU Zhuoran1; FU Qiang2; （1.School of Public Administration,Dongbei University of Finance and Economics,Dalian 116025,China;2.College of Teansport & Communications,Shanghai 201306,China）

Keywords:: grain:maritime risk; risk assessment; multi-criteria decision-making; failure modes and effects analysis (FMEA); fuzzy best-worst method (FBWM); Bayesian network

摘要:: 针对中国粮食进出口海运供应链多源不确定性，综合失效模式与影响分析方法（failure modes and effects analysis，FMEA）、模糊最佳最差方法 (fuzzy best-worst method，FBWM)和规则贝叶斯网络（rule-based Bayesian network，RBN），提出一套基于FMEA-FBWM-RBN的风险评估框架，实现关键环节风险识别与情景化推断。首先，采用FMEA，从极端天气、运输堵塞、通道安全、政治与法律、货物与船舶5个方面，系统识别出19种典型失效模式；然后，引入FBWM，量化专家评估的主观性与模糊性；最后，构建三级规则贝叶斯网络，在GeNIe软件平台实现基于贝叶斯网络的因果传播与情景推断。通过实证分析，该方法能识别各失效模式及失效类别的风险清晰值，并据此筛选出影响粮食海运安全的关键风险因素为粮食货物属性、运河与海峡的政策调整、出口国政策限制与出口禁令。通过敏感性分析，验证了模型的稳健性与解释性。

Abstract:: This study proposes an integrated risk-assessment framework based on FMEA-FBWM-RBN to address multi-source uncertainties in China’s seaborne grain import-export supply chain,enabling key-node risk identification,consistency-aligned weighting,and scenario-based inference to assist resilience building and policy preparedness.It systematically identify 19 failure modes across five dimensions,namely,extreme weather,traffic congestion,corridor security,politics & law,and cargo & ships;use the fuzzy best-worst method (FBWM) to aggregate expert judgments and handle subjectivity and vagueness,and implement causal propagation and scenario inference via a risk Bayesian network (RBN) in GeNIe.Empirical analysis yields the crisp risk values (CV) for each failure mode and category,from which the key drivers affecting grain maritime safety are identified,including the nature of grain,policy adjustments in canals and straits,as well as export country policy restrictions and bans,etc.Sensitivity analysis further validates the model’s robustness and interpretability.