ThailandReal estate2023 – ongoingAsset World Corp (AWC)
AI item matching for bulk hotel purchasing
AWC cut a 200,000-line purchase order item list by about 65% by matching similar items across its hotels with AI
Problem
AWC's hotels bought many near-identical items under different descriptions. Its raw purchase order data held more than 200,000 SKUs, which made bulk purchasing hard.
Approach
AWC built a procurement platform that uses natural language processing to find similar items across its affiliated hotels. It combined the model with a cloud data lake and dashboards for faster queries. AWC said it planned to extend the model to more categories and to supplier and inventory work.
Results
Takeaway
Cleaning up item descriptions was the first step, and the saving AWC reported came from the bulk purchasing it made possible.
Categories
- Size
- Large enterprise
- Industry
- Real estate
- Business function
- Logistics and supply chainProduction and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
AWC's release on winning an IDC Future Enterprise Award. It describes an NLP-based procurement platform that matches similar items across its hotels. It reports a SKU reduction of about 65%, human input cut from months to weeks, and a 20% saving from bulk purchasing.
Full details are in the original sources.
