ChinaFood and restaurants2026 – ongoingFoshan Haitian Flavouring and Food
AI soybean inspection and aroma assessment
Haitian used AI to screen 2.5 trillion soybeans annually and assess fermentation aromas
Problem
Haitian needed to inspect individual soybeans at production scale and make experienced workers' aroma assessments reproducible.
Approach
Haitian developed food-specific algorithms using its accumulated quality and taste data. Its Gaoming factory used image recognition to screen soybeans and an AI electronic nose backed by an aroma database.
Results
Takeaway
The systems addressed ingredient screening and sensory assessment separately. Haitian did not disclose rejection accuracy or a measured reduction in quality failures.
Categories
- Size
- Large enterprise
- Industry
- Food and restaurants
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer visionPredictive analytics and ML
- Adoption
- Built in-house
- Goal
- Quality and risk
- Evidence
- Company's own claim
Sources
Source summaries
The local government publication described a media visit to Haitian's factory and quoted its digital engineering chief. It reported annual screening of 2.5 trillion soybeans and an electronic nose using an aroma database.
Full details are in the original sources.
