TaiwanManufacturing2019 – ongoingChung Hwa Pulp
Automatic virtual metrology for paper quality at the Jiutang mill
Chung Hwa Pulp's mill manager reported paper machine uptime rising from about 80% to 94% after virtual metrology
Problem
Checking paper quality meant stopping lines and taking manual samples. Problems in the coating process were found late.
Approach
The Jiutang mill applied automatic virtual metrology, developed with National Cheng Kung University, to its coating process. The system predicts paper quality from machine data, so all product is checked online. Staff watch deviations on a control-room platform and supervisors get LINE alerts. Of more than 1,400 measurement factors, over 500 were chosen for the AI analysis.
Results
Takeaway
Predicting quality from existing machine data replaced stoppages for sampling. The figures come from the mill manager, not an outside audit.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Partner co-development
- Evidence
- Company's own claim
Sources
Source summaries
Global Views Monthly described how Chung Hwa Pulp and Yuen Foong Yu use AI. At the Jiutang mill, automatic virtual metrology built with National Cheng Kung University predicts paper quality. The mill manager gave the efficiency, uptime and CO2 figures.
Full details are in the original sources.