| I&T Solution |
AI-assisted data submission analysis system for the Mandatory Energy Efficiency Labelling Scheme (MEELS)
(REF: S-2105) |
| Trial Project |
|
| Solution Feature |
- Automate data entry and vetting process
- Process application forms, lab reports, certificates, handwritten, company chops, and signatures from PDFs, images, or hardcopy
- Orchestrates specialised AI agents for routing, printed / handwritten text extraction, stamp / signature verification, database cross-checking, and report drafting
- Achieve ≥95% overall accuracy and >98% for critical parameters
- Self-learning from manual corrections for continuously improvement and reducing manual intervention and errors
|
| Trial Application and Expected Outcome |
- Conduct trial using existing submissions of mixed quality, handwritten entries, and chops to compare AI-extracted data against manual entry for accuracy, speed, and error rates
- Provide real-world cases such as poor scans and varied handwriting to train the self-learning module
- Manual correction fine-tunes models to address limited historical datasets
- Deploy the trial in parallel with existing workflow where AI flags uncertainties for human review, with corrections captured to update the system for ongoing improvement
- Trial validates integration databases for real-time certificate checking and proves automated drafting of report
|
| Additional Solution Information |
AI-assisted data submission analysis system for MEELS_v0.pdf
|
| Info on I&T Solution Provider |
| Solution Provider | : | Hong Kong Productivity Council | | Address | : | HKPC Building, 78 Tat Chee Avenue, Kowloon, Hong Kong | | Contact Person | : | CHAN Shuk Ching |
| Position | : | Senior Consultant, Emerging Applications | | Tel | : | (852) 2788 5470 | | Email | : |
scchan@hkpc.org |
|
