Conference

We’re speaking at WTC Montreal

May 17–21

Catch Differing Site Conditions as begin to be encountered, not after the project ends.

The sooner variance from the GBR is identified, the easier it is to resolve. GroundedAI turns field observations into structured data that's easy to compare against the GBR, so a differing site condition shows up as it's encountered, not when a claim needs to be assembled.

The gap today

Differing Site Conditions are usually spotted twice

Once informally in the field when someone says "this doesn't look like what we were expecting," and again months later when a claims manager assembles the evidence. Between those two moments, the record becomes fragmented. Photos on different phones, notes in different books, observations in Teams chat threads.

The Workflow Stages

What changes when early detection is built into the workflow (compressed)

Three stages. Two products. One connected workflow from face to decision.

STAGE 1 - CAPTURE - LITHOS

Capture what's happening underground

Underground conditions recorded in 3D with rock mass classifications and photographic evidence, time-stamped at the point of observation.

STAGE 2 - COMPARE - STRATA

Compare against the baseline

Strata enables easier comparison of GBR predictions, design assumptions, or modelled conditions overlaid against current captures.

STAGE 3 - FLAG - STRATA

Flag and act

When conditions diverge, the record is already generated. Variances are documented with a specific date and location.

The Platform

The product that drives this outcome (Strata-led)

Strata

Where plan meets reality

Capturing underground conditions is the starting point, but identifying a Differing Site Condition requires monitoring of ongoing datasets of what was predicted vs.what was encountered. Strata structures data so that it's easy to compare baseline data (GBR, forecasts, model predictions) and current datasets (from Lithos).

Time and location-based ground records enable early identification of differing site conditions based on real, visible data.

What you get

What you have to show for it

Differing Site Conditions flagged as they are encountered, not months later. Time-stamped photographic and 3D datasets of the variance, tied to the exact location. A dataset that can be easily compared against the GBR, regardless of who is capturing the data. An early-detection record that promotes proactive mitigation and strengthens contemporaneous documentation.

Frequently asked questions.

Data captured in Lithos flows directly into the Grounded platform, where teams can review conditions, generate reports, and align on decisions.

How is this different from Build Defensible DSC Records?
What is a GBR?
Can other teams, an owner, or subcontractors access the data on Strata?
Related Workflows

Used across key workflows.

See how structural mapping connects to the broader GroundedAI platform.

See how differing site conditions get flagged at as they are encountered.

We'll walk through the capture-to-comparison workflow with a real tunnelling scenario.