Tobacco Asia

July/August 2025 - Vol 29-No 4

Issue link: https://read.dmtmag.com/i/1537288

Contents of this Issue

Navigation

Page 74 of 83

Tobacco Asia 75 tem performance while controlling information technology construction costs. China Tobacco (Guangxi) Commercial Corporation went further, constructing a trinity AI infrastructure of "com- puting power + data + models" following the path of "unified system architecture, unified computing power foundation, and coordinated data supply." This provides solid ground for various intelligent applications. Comprehensive deployment and deepening application phase Currently, numerous branches under CNTC na- tionwide have initiated AI localization deployment efforts. By building intelligent platforms covering the entire industrial chain, they aim to achieve da- ta-driven, precise decision-making. These applica- tions can be categorized into three types based on their value proposition: 1.Operational innovators: For instance, the AI visual inspection system at Changde Tobacco Factory under China Tobacco (Hunan) Industrial Corporation achieved a 99.79% recognition accura- cy rate for moldy tobacco leaves while the "Rolling and Packaging Maintenance Knowledge Assistant" at Wuhan Tobacco Factory under China Tobacco (Hubei) Industrial Corporation achieved second- level fault diagnosis response by analyzing 13,000 maintenance records. These applications directly address pain points in quality control and equip- ment maintenance during the production process, transforming the experience of veteran workers into reusable digital assets. 2.Decision-making enablers: For exam- ple, the "Guiyan AI Assistant" of China Tobacco (Guangxi) Industrial Corporation covers five ma- jor scenarios including intelligent q&a, smart data querying, and compliance review. Particularly, its marketing data analysis report function provides scientific decision-making basis for management through in-depth mining of 58 key indicators. The "Anyang AI Assistant" at Anyang Tobacco Factory under China Tobacco (Henan) Industri- al Corporation has built a local legal knowledge graph, establishing an intelligent risk early-warning mechanism in high-risk areas such as procurement and engineering. 3.Business model pioneers: For instance, the "Xiaotao Lawyer 2.0" of China Tobacco (An- hui) Commercial Corporation redefines the form of legal education, providing personalized legal services through voice interaction and contextual understanding, while the "Lin Xiao'an" AI assis- tant of Linhai branch under China Tobacco (Zhe- jiang) Commercial Corporation has restructured the safety management process, achieving closed- loop management of inspection, reporting and rectification. These innovations not only enhance work efficiency but also give rise to new business models. Particularly noteworthy is the Automatic Tobacco Bale Scanning System developed by China Tobacco (Hunan) Leaf Redrying Company. It overcame the technical challenges of dynamic scanning and built the industry's first tobacco bale and label morphology database. This innovative AI technology is transforming China's tobacco industry from a "manufacturing factory" into a "smart enterprise."

Articles in this issue

Archives of this issue

view archives of Tobacco Asia - July/August 2025 - Vol 29-No 4