HORUS Intelligence
Decision Twin
Simulate the decision before you make it
Stochastic programming on your real operational data: 7 agents run DES + Monte-Carlo over bed logs, OR costing and claims facts to de-risk capacity, staffing, supply and scheme decisions. Outputs are population-level only — the device line is a hard invariant.
Open step-down unit (12→10 beds, budget-capped)
optimizedRenegotiate NHSO scheme — denial rate 18%→9%
4 sign-offs pendingIn-source blood supply — 3 scenario runs
decidedHITL gate · 4 sign-offs release optimized → decided
Capabilities
7 agents: Demand, Capacity, Workforce, Supply, Reimbursement, Quality, Compliance
Initiative Workbench (Capture → Analyze → Model → Guide) + Portfolio Command
4-role HITL gate releases optimized → decided — the SaMD firewall
Calibrated by live twin_metric_* RPCs; fail-soft to seed with isLive badge
Rule-row constraints: budget caps prune recommendations (12 beds → 10)
Inside the twin
Distributions, not point estimates
Every initiative runs as thousands of discrete-event + Monte-Carlo simulations on your live operational data. The twin shows you the spread of outcomes — and the gate that turns a run into a decision.
Distribution of outcomes, not a point estimate. The capacity line sits above P95 — the step-down plan holds at the 95th-percentile surge. Calibrated by live twin_metric_* RPCs.
Forecast mean with 90% band. Week-8 demand crosses staffed beds — the Workforce agent proposes a roster shift before the breach, not after.
Every option scored on the same axes. Budget-cap rule rows prune infeasible runs (12 beds → 10) — the twin never recommends what the rule engine forbids.
optimized → decided
LLM writes commentary, never decides a number