Home / Solutions / Operations Advisor

Operations
Advisor

Production deployed

Product maturity · Data and deployment requirements

Threshold alerting needs somebody to know the number in advance. Most of what quietly costs an operation money has no number: a haul route that got slower, a drill that dropped off its own average, a face that stopped behaving like itself. This compares every unit of work against its own history and writes up the departures, with the effect size, the sample it is standing on, and how confident it is.

Talk about your operation How the baseline works

In practice

What an advisory looks like.

Two findings on the same haul route, at different points in the shift. Nobody configured a limit for either. The route was measured against its own median across several hundred prior cycles, and the wording, colour and confidence all move with the size of the departure.

An amber Attention advisory: a route's median cycle time at 25.6 minutes against a baseline median of 24.4 minutes across 780 prior cycles, a 5 percent increase, with queue and delay 6.5 minutes above baseline, a suggested inspection, and 85 percent confidence

A red Action advisory on the same route: median cycle time at 27.8 minutes against a baseline median of 24.3 minutes across 763 prior cycles, a 14.5 percent increase, with queue and delay per cycle up 2.5 minutes, a suggested inspection, and 90 percent confidence

What to notice: a second signal has to agree before either card appears, queue and delay time on the same route, which is what separates a genuinely slower haul from ordinary variation. The sample count is on the face of the card so a supervisor can weigh it in seconds. And the suggestion names what would make it a false alarm, a recent routing or dispatch change. An advisory that tells you how it could be wrong is one people will act on.

And that they are graded: five percent is amber, headed Attention, at 85 percent confidence. Fourteen and a half percent is red, headed Action, at 90 percent. Small drift gets noted rather than escalated, so a channel stays triageable by scrolling instead of reading.

Screenshots from production systems. Site identifiers, serial numbers and personnel details have been altered. Public demos use synthetic data.

The baseline

Every deployment starts quiet. That is the point.

The advisor cannot report anything until it knows what your operation's normal is, and a baseline is not something that can be shipped in a box. It has to be accumulated from your own completed work. Several hundred cycles on a route before a comparison is worth making, which a busy haul reaches in days and a road to a rarely used dump takes considerably longer. Through that period the service is running and building history, and it says nothing.

This is worth being direct about, because it is the honest test of any advisory system sold into mining. Anything that starts making confident claims about your operation in its first week is not comparing your operation to itself. It is comparing you to a model built somewhere else, on somebody else's pit, somebody else's fleet, somebody else's haul profile. The waiting is what makes the output about you.

Phase 01

Accumulate

The service reads completed work and builds a per-unit history. Nothing is published and nobody is notified. If your warehouse already holds cycle history, this phase is largely already done and the runway shortens accordingly.

Phase 02

Calibrate

Findings are generated but held back for review rather than posted to a channel. You see what it would have said, and the sensitivity gets set against what your supervisors actually found when they went and looked.

Phase 03

Publish

Advisories start arriving in the channel, graded. The baseline keeps moving from here, so the reference is always recent history rather than a number frozen at commissioning.

Ongoing

The Baseline Never Stops Moving

Because it is recomputed rather than fixed, the comparison stays valid as the pit deepens, routes change and the fleet turns over. A threshold set at commissioning is wrong within a quarter and nobody notices.

Mechanism

Why the findings hold up in a meeting.

Baseline learnedEach unit of work is compared against its own recent history rather than a number somebody set once and nobody revisited.
CorroboratedA deviation on one measure is a curiosity. A second measure has to agree before it takes anyone's attention.
Effect and sampleEvery advisory states the size of the change and how many prior observations it rests on, so it can be weighed rather than trusted.
Written as an actionThe headline is an instruction, not an observation, and it names where to look and what to rule out first.
Confidence statedSomething that never admits doubt gets either over-trusted or ignored, and both are worse than being told how sure it is.
Severity gradedWording, colour and confidence scale with the departure, so a channel stays readable at a glance.
Operational decision supportIt recommends an inspection. It does not move trucks, change destinations, or touch your fleet management system.

Scope

Haul routes are the first case, not the only one.

Integrates through a validated warehouse

The cards above happen to read a Cat Minestar dataset, but the method needs only a history of completed work with a start, an end, a duration, and somewhere the time went. Availability and semantics vary by fleet system and approved export. Once those records are in your own warehouse, the advisor reads the warehouse rather than any vendor's schema, with interface changes assessed and revalidated when a fleet system changes.

Wenco · Jigsaw · Modular · Minestar

Any Unit Of Work With A History

Drill patterns against their own metres per hour. Crusher throughput by feed source. Loading tool performance on a given face. Plant recovery by ore type. Pump runtimes, tyre life by route, fuel burn by machine and grade. Wherever an operation has a normal it never wrote down, there is something here to compare against.

Same engine, different unit of work

Built On The Warehouse

The advisor is a consumer of the governed warehouse rather than a separate stack. If the shift calendar, the rehandle logic and the coordinate transforms are already settled there, the advisory inherits definitions everyone already agrees on.

Pairs with production reporting
See the warehouse →

Production deployed

A learned model alongside the baseline advisor.

The learned machine-learning model runs in production alongside the statistical baseline advisor. It adds pattern analysis across multiple variables while the baseline provides the historical comparisons behind an advisory.

Teams can review the operating context, sample counts and comparisons when investigating a finding. The model and baseline work together to support traceable operational decisions.

StatusThe learned model and baseline advisor are complete and production deployed.
RelationshipThe model informs and ranks findings alongside the explainable baseline comparisons.
DataRepresentative operating history, agreed definitions and feedback on investigated advisories support configuration and evaluation for each mine.

Next step

What is slowly getting worse that nobody has noticed?

If you can answer that, you did not need this. The interesting version of the question is how you would know, and that is usually the whole conversation. Tell us what history your systems keep and the runway becomes obvious quickly.

Start a conversation Tell us what you keep history on Threshold alerting The other half of the channel

Location Timmins, Ontario, Canada

Automation & on-premises AI

Scope the capability before deployment.

A route or machine can drift without crossing a fixed alarm threshold.

Built and running in mine operations. The implementation examples show the workflow and outputs. The Technology & Data Assessment defines the data connections, configuration and acceptance criteria for your mine. Public demonstrations use synthetic data.

What it doesBaseline comparisons with sample counts, effect sizes and confidence; a learned machine-learning model; local inference infrastructure; autonomous operational AI agents; advisory and document workflows.
Data requiredValidated cycle history, route/machine identifiers and operating context; approved documents for retrieval workflows.
PrerequisitesRepresentative history, baseline calibration, false-positive review, access controls, configured agent permissions and an operational decision owner. Tooth-and-shroud AI monitoring is the only proof of concept; see the product maturity legend.
OutputsTraceable advisory cards, inspection suggestions, evaluation records and operational workflow outputs under configured permissions and review.
ImplementationAccess, integration, cybersecurity and data quality are assessed before duration or acceptance is committed. Existing workflow descriptions are not delivery guarantees.

Demo coverage and limits

Commercial engagement

Software subscriptions provide platform access and include applicable updates, defect fixes and standard product maintenance. Optional Premium Support is $50,000 CAD per mine annually for additional operational assistance under agreed service terms. Custom development, new integrations, custom reporting, specialized analysis and advanced troubleshooting outside support scope are separately billed consulting at $260 CAD/hour, unless expressly included in a signed agreement. All amounts exclude applicable taxes.

Implementation is quoted for deployment requirements. No consulting hours or Premium Support are automatically bundled with software. Estimate software, support and consulting.

Technology & Data Assessment Full pricing and service boundaries →

OperationsAdvisor