۱- Introduction

۲- The risk early warning method

۳- Case study

۴- Discussion

۵- Conclusion

References

Abstract

Food safety risks has received great attention in all world. And the reasonable effectiveness of security warnings can reduce public panic and risk losses. Therefore, this paper proposes an improved risk early warning method for food safety detection data based on the analytic hierarchy process (AHP) integrating the quality control analysis method. The AHP based on the entropy weight can obtain risk values for food safety component data. And the risk matrix of the risk component is obtained by the risk probabilities of the components. Then the corresponding risk levels are calculated using the quality control analysis method to release the risk warning information. Finally, a case study of dairy product safety data from the GuiZhou province in China is conducted to verify the feasibility and reliability of the proposed method. Moreover, the proposed method can scientifically and reasonably determine the risk level information. Furthermore, the risk management is provided to effectively reduce risk losses of the country though relevant quality inspection departments.

Introduction

With the rapid development of the economy, the food safety and quality have raised a higher requirement. If making correct and timely warning of food safety, people’s fears will be alleviated, and the harm caused by the food security crisis will be reduced. Nowadays, there are more panic and unintended consequences bring by false warnings. And the food safety risk is serious. Meanwhile, more and more food safety problems involving complex food safety data are occurred (Ma, Hou, Liu, & Xue, 2016). Because the food safety risk monitoring foundation of China is weak, it is very important to customize a risk monitoring model based on the basic national conditions (Tang, 2013). Due to the superior processing characteristics of complex food safety data technology, many dig data analysis and artificial intelligence methods of food safety risk assessment and early warning were proposed (Liu, Li, Yang, & Guo, 2018a; Wang, Yang, Luo, He, & Tan, 2015). Samuel et al. (Samuel, Asogbon, Sangaiah, Fang, & Li, 2017) used the fuzzy analysis hierarchical process (AHP) technique to calculate the global weight of attributes based on their individual contributions and predicted the high frequency risk of patients by training the artificial neural network (ANN) classifier. Wang et al. (Wang & Yue, 2017) formulated an early warning strategy for the safety risks arising from food transportation in the real-time monitoring of food safety to reduce the risk of food supply chain.