United StatesEnergy and utilities2024 – ongoingOccidental Petroleum
AI classification and metadata extraction for land-lease documents
Occidental's automated lease-document review saved $4 million and 12 months, MongoDB reported
Problem
Occidental needed to classify a batch of 1.5 million land-lease documents and extract information. It had planned to hire contractors for manual review.
Approach
Its engineers built a cloud processing pipeline with MongoDB Atlas, AWS Lambda and Azure AI Document Intelligence. OCR extracted text, and a language model classified documents into 140 categories and extracted metadata. Stateful records allowed processing to resume after failures.
Results
Takeaway
The reported savings concerned a defined document-processing assignment. Additional savings from other applications remained forecasts.
Categories
- Size
- Large enterprise
- Industry
- Energy and utilities
- Business function
- Internal work and knowledge managementFinance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Generative AIComputer vision
- Adoption
- Built in-house
- Evidence
- Vendor case study
Sources
Source summaries
MongoDB described Occidental's automated lease-document pipeline and cited its engineers' presentation at MongoDB .local Austin 2024. It reported $4 million and 12 months saved.
Full details are in the original sources.
