Meana Petro
DF-02FoundationalWave 1AIDomain

Structured Extraction from Reports

In one sentence

Pulls the numbers out of the reports — depths, rates, pressures, volumes, dates — and puts them into a database with a link back to the page they came from.

The problem

Searchable text is a large improvement over scanned images and it is still not data. Nobody can plot a trend, build a model or run a query against prose. The numbers have to become rows.

How it works

Document type is classified first, then a type-specific extraction schema is applied — a daily drilling report, a PVT study and a well test report each have their own. Extraction combines layout-aware models with domain validation rules: a depth that decreases, a rate outside physical range or a date out of sequence is rejected rather than stored. Every extracted value keeps a citation to its source page and coordinates.

Data required

  • OCR output from the document pipeline
  • A defined target schema per document type, agreed with the domain team
  • Validation rules and physical ranges for each field
  • A labelled sample per document type, typically fifty to two hundred documents

What you get

  • Structured records in the relational store with full source citation
  • Extraction confidence per field with a review queue for low-confidence values
  • Coverage report: how many documents of each type were processed and what share of fields were populated

What it cannot do

Each new document type needs its own schema and labelled examples — this is not a single model that reads everything. A document type with only a handful of examples in the archive is not worth automating, and we say so rather than delivering a fragile extractor. Values that were wrong in the original report are extracted faithfully wrong; validation catches the physically impossible, not the merely mistaken.

Every capability page states a real limitation. If a vendor cannot tell you where their product stops, you are not being told the truth about where it starts.

Related capabilities

  • DF-01Foundational

    Field Document OCR Pipeline

    Turns decades of scanned reports, logs and maps into searchable text, including Persian and Arabic documents and handwritten annotations.

    Data FoundationAI
  • DF-07Foundational

    Automated Data Quality Control

    Continuously checks incoming data for stuck sensors, impossible values, missing periods and broken units, and reports the field’s data health as a number.

    Data FoundationITAI
  • DR-13Foundational

    NPT Root-Cause & Offset Well Lookup

    Reads every daily drilling report the operator has ever filed and answers, in seconds, what went wrong at this depth in this field before.

    Drilling IntelligenceAI
  • SR-13Foundational

    PVT Modeling & EOS Tuning

    Extracts fluid study data from decades of laboratory reports and tunes the equation of state against all of it, not against the one report someone had to hand.

    Subsurface & Reservoir IntelligenceDomainAI
  • KC-01Foundational

    Technical Knowledge Assistant (RAG)

    Ask a question about your own field in Persian, Arabic, English or Spanish and get an answer with the report and page it came from.

    Knowledge & CopilotAI