Journal of Networking Technology Yingchun Guo, Jianyue Ma, Shubo Zhang 15 4 2024 https://doi.org/10.6025/jnt/2024/15/4/142-149 https://www.dline.info/jnt/fulltext/v15n4/jntv15n4_4.pdf Food safety management is crucial for social stability and is an important component of the country’s long-term development strategy. Previous food safety management methods failed to achieve dynamic supervision of various links in the food production-to-sales process and needed more objectivity in food safety testing. Therefore, this paper introduces the Markov model into food safety management and optimizes it using the cuckoo search algorithm to improve parameter convergence speed and solve the global optimal solution problem. Experimental results demonstrate that based on the improved hidden Markov model, the food safety management model can effectively achieve dynamic food safety supervision and improve food safety risk assessment accuracy. In practical applications, it significantly enhances food management efficiency and improves technical and scale efficiency.