多维度食品安全评估指数体系模型的构建及应用

目的:通过食品安全评价指数体系来评价地区食品安全状况。方法:在构建的指数体系中，综合考虑影响食品安全的各种因素，如地市规模、食品种类、检测项目危害、生产地等，并采用多种数据来源对模型进行支撑。结果:利用2021年广西食品安全评价抽检数据进行实证，分别使用经典大样本估计法和经验贝叶斯估计法进行模型计算后，发现基于食品种类的安全指数的评估结果与简单合格率评估结果存在一定差异，但两种算法均显示餐饮食品的安全情况显著低于其他食品大类。此外，使用贝叶斯估计方法有效解决了合格率为100%或数值普遍接近时的计算问题。结论:该模型可以实现从地区、食品种类、地区和食品种类等多角度、不同层次计算得到食品安全指数结果。

Construction and application of multi-dimensional food safety evaluation index system model

Objective: It was the method of evaluating regional food safety status through food safety evaluation index model. Methods: In the index model, various factors affecting food safety, such as the size of the city, the types of food, the hazards of the test items, the place of production, etc., were comprehensively considered, and a variety of data sources were used to support the model. Using this model, the results of food safety index could be calculated from multiple angles and different levels, such as region, food category, region and food category. Results: Sampling data of Guangxi food safety assessment in 2021 had been utilized for empirical research. The classic large sample estimation method and empirical Bayesian estimation method had been applied for model calculation. It was found that the results from these two methods had a certain differences, but both methods got the same results showing that restaurant and catering food safety was significantly poorer than other food categories. In addition, the Bayesian estimation method could effectively solve the calculation problem when the pass rate was 100% or the values were largely close. Conclusion: The model could calculate the food safety index results from multiple angles and different levels, such as regions, food types.