Abstract: Product quality safety problem of logistic units in production supply chains is studied.Quality control requirements of traceability systems are analyzed.A pre-warning model that can be applied to supply chain traceability systems is proposed.In the model，all of detection data existing in the whole supply chain are analyzed.Reasons that may cause logistic unit quality safety problem are diagnosed and pre-warned.This model focused on four abnormal data types in supply chain，mathematical algorithms for the abnormalities are developed by combining Radial Basis Function（RBF） neural network and statistical analysis methods.The simulation results show that the proposed pre-warning model can identify abnormal data types effectively.Meanwhile，the detection data can be monitored and pre-warned in time as well.

摘要：

针对物流单元在生产供应链中出现的产品质量安全问题，分析可追溯体系中的质量监控需求，提出了一种用于供应链可追溯系统的预警模型，对整个供应链的检测数据进行汇总分析，预警可能导致物流单元质量问题的原因并进行诊断。针对供应链中四种常见的数据异常，提出了基于径向基函数神经网络、统计学分析的分析算法。实例分析和结果表明，该预警模型能够有效地区分异常数据的类型，实时地对追溯系统的检测数据进行监控及预警。