Risk Engine
Risk is a number. We show you how we got there.
We combine asset condition, maintenance history, weather exposure, and system dependence into one score. Built to be explained, not trusted blindly.
9 scoring layers → compound risk score
Every score should answer why.
Teams should not have to choose between better prediction and better trust. The risk engine supports both — driver ranking, uncertainty context, and full calculation traceability.
Compound risk score
Contributing factors — driver breakdown
Driver ranking shows the strongest contributors so teams know what is pushing urgency up. Confidence bounds communicate where the model is strong and where more data would improve precision.
Dependency-aware risk intelligence.
Traditional scoring treats every asset in isolation. A chiller is scored by its age and condition — nothing more. But a chiller that feeds an OR wing, shares a cooling loop with a data closet, and depends on a generator that is already flagged — that is a different risk entirely.
Rivolq uses graph-based modeling to map every dependency chain in your facility — which systems feed which spaces, which assets share infrastructure, and where a single failure cascades into compound disruption. The risk engine scores the relationship, not just the asset.
Blast radius mapping
See which spaces, systems, and operations a single asset failure would disrupt
Cascade detection
Identify chains where deferred maintenance on one asset amplifies risk in connected systems
Dependency-weighted scoring
Assets that more systems depend on are scored with higher consequence multipliers
Graph-backed capital sequencing
Prioritize replacements that reduce the most connected risk, not just the worst individual score
Hover any node to trace what depends on it.
Exposure $2.1M
Built for decisions that have to hold up outside operations.
The risk engine is not only for analysts. It is for the moments when facilities, finance, and leadership all need to understand the same recommendation.
Full calculation traceability.
Every score can be traced back to data sources, assumptions, and logic layers. The path from input to output stays inspectable, which matters when recommendations are used in governance or budget settings.
Designed for the room, not just the dashboard.
Model outputs support executive review, not just internal dashboard use. Explainability is treated as a product requirement, not a post-processing add-on.
Reproducible logic, defensible results.
The engine is built around reproducible logic so the same environment yields a defensible result instead of a shifting story nobody can verify. Better decisions under pressure, not opaque model theater.
The risk engine becomes most valuable when multiple stakeholders need the same answer from the same signal.
Facilities teams
“See what deserves urgency before the week disappears into noise.”
The engine helps operators separate active disruption from the assets that are quietly becoming much more expensive to ignore.
Finance & capital planning
“Understand why one project should move before another.”
Risk scoring becomes most useful when it helps compare timing, exposure, and consequence in a way that funding conversations can actually use.
Executive review
“Get a cleaner explanation than "this system feels urgent."”
Leadership can see why a recommendation exists, what is amplifying the risk, and what kind of delay would likely make the outcome more expensive.
See the risk engine with your facility type, not just a generic demo.
We'll walk through how the scoring logic works, what kinds of signals show up in practice, and how the output flows into reporting or capital planning.