Meana Petro
DR-03FrontierWave 3AIDomain

Permeability from Mud Loss Data

In one sentence

Mud that disappears into the formation is telling you something about how easily fluid moves through the rock. This reads that signal instead of discarding it.

The problem

Loss events are recorded as operational incidents and filed. The hydraulic information in them — rate, pressure, duration, depth — is rarely used quantitatively.

In naturally fractured carbonate fields, where permeability varies by orders of magnitude over short distances, that is a large amount of free subsurface data being thrown away.

How it works

Each loss event is inverted against a radial-flow model with the observed ECD, loss rate and exposure time to produce a permeability-thickness bracket rather than a single value. Results are reconciled against core, well test and log-derived permeability where those exist.

Output is delivered as an uncertainty range per depth interval, feeding the geomodel as soft data with an explicit confidence weight.

Data required

  • Daily drilling reports and mud reports with loss volumes, rates and depths
  • Real-time ECD or a calibrated hydraulics model to reconstruct it
  • Openhole logs over the loss intervals
  • Any core or well-test permeability in the field for calibration

What you get

  • Permeability-thickness estimates with uncertainty bands, by depth interval
  • A field-wide loss-event atlas linked to structural and facies maps
  • RESQML-compatible soft-data export for the geomodelling team

What it cannot do

This is frontier work and we say so. The inversion is non-unique: a large loss into one open fracture and a moderate loss into a high-permeability matrix can produce the same signature. It is useful as a trend and as soft conditioning data, and it is not a substitute for a well test or a core plug.

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

  • DR-06Foundational

    Real-Time Torque, Drag & Hydraulics

    Compares what the string should weigh and how hard the pumps should work against what is actually measured, and flags the gap.

    Drilling IntelligenceDomainAI
  • SR-07Foundational

    Well Log QC & Missing Log Synthesis

    Finds the bad intervals in the log archive and reconstructs curves that were never run, so the field has a complete set to work from.

    Subsurface & Reservoir IntelligenceAI
  • DR-09Frontier

    AI-Assisted Geosteering

    Keeps the horizontal well inside the target layer by continuously re-matching the measured logs to the geological model.

    Drilling IntelligenceAIDomain
  • DR-11Frontier

    Automated Drilling Program Design

    Drafts the casing, mud and BHA programme for a new well from the field’s own drilling history, for an engineer to review.

    Drilling IntelligenceAIDomain
  • DR-15Frontier

    Managed Pressure Drilling Support

    Helps hold bottom-hole pressure inside a narrow window by predicting what the choke needs to do before the transient arrives.

    Drilling IntelligenceDomainAI