Skip to content
SIH Buddyby Ganeev Singh
Dev
All problem statements
SIH26121Strong pickacceptance 4/5

eRTMAC-NWIS (Nearby Wells Intelligence System): An AI-Powered Offset Well Knowledge and Decision Support Platform for Drilling Operations

Oil India Limited · Smart Automation · Software

One of the better-specified industry problems in this block — build on the public Volve or Norwegian well reports, make the depth-triggered alert deterministic, and be candid that the corpus is foreign.

What it actually is

When a drilling crew hits trouble at a certain depth, the odds are a nearby well hit the same trouble years ago — but that knowledge sits in old completion reports and in the memory of whoever was on site. The ask is a platform that reads those historical documents, puts the nearby wells on a map, and warns the active crew before they reach a depth where neighbouring wells ran into problems.

What to build

A decision-support platform that ingests historical well completion and daily drilling reports through OCR and information extraction, structuring the events that matter — mud losses, kicks, stuck pipe, casing programmes, cementing practice, formation tops — into a queryable per-well record; a map showing offset wells within a radius of the active well with their event history; a correlation view aligning drilling parameters and formation markers across neighbouring wells on a common depth axis; and a proactive alert that fires as the active bit approaches a depth or formation where nearby wells logged an incident, naming which well, what happened and at what depth.

Smallest thing that wins the room

Advance the active well's bit depth on the timeline and watch an alert fire — three offset wells within two kilometres logged mud losses entering this same formation, each named with its depth and what the crew did about it.

How crowded this one gets

A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.

Moderate85–190 teams expectedroughly 1 in 69–161 wins it

Quieter than 72% of the 226 · #63 of 226 by expected field

A normal-sized field. Your idea has to be good, not miraculous.

Why: company-sponsored statements drew the smallest fields of all.

This is a guess, not a fact

Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.

The scores

The number is the shorthand. The line under it is the reason.

What you will be writing

  • PaddleOCR / docling for scanned report ingestion
  • LayoutLM or LLM-based event extraction
  • PostGIS offset-well spatial queries
  • Depth-indexed time series (TimescaleDB)
  • React + Mapbox well map with depth correlation panel
  • Rule-based proximity alerting engine
  • Drilling operations
  • Document intelligence
  • Geospatial decision support

Prior art to read before you start

offset well knowledge retrieval · operational event extraction from reports · depth-triggered proactive alerting

Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.