Forecast alert fatigue and missed risk loss

What it's for

Quantifies whether 'manage alerts, not noise' is working by forecasting duplicate storms, attention saturation, missed risks and their cash tail before buyers add more rules.

What you give it

Inputs split into evidence read from your connected systems, calibration your team owns, and numerical controls that affect precision but never the result's meaning.

{
  "attention_states": [
    {
      "acknowledgement_prior_alpha": 9,
      "acknowledgement_prior_beta": 1,
      "alert_capacity_per_period": 12,
      "evidence_verified": true,
      "fatigue_rank": 0,
      "id": "fresh",
      "timely_action_prior_alpha": 8,
      "timely_action_prior_beta": 2,
      "transition_prior_weights": [
        8,
        2
      ]
    },
    {
      "acknowledgement_prior_alpha": 6,
      "acknowledgement_prior_beta": 4,
      "alert_capacity_per_period": 8,
      "evidence_verified": true,
      "fatigue_rank": 1,
      "id": "loaded",
      "timely_action_prior_alpha": 5,
      "timely_action_prior_beta": 5,
      "transition_prior_weights": [
        2,
        8
      ]
    }
  ],
  "current_alert_exposures": [
    {
      "alert_class_id": "delivery-risk",
      "alert_sensitivity": 0.8,
      "condition_rate_prior_rate": 100,
      "condition_rate_prior_shape": 2,
      "duplicate_probability_prior_alpha": 1,
      "duplicate_probability_prior_beta": 9,
      "evidence_verified": true,
      "false_alert_rate_per_opportunity": 0.02,
      "forecast_condition_opportunities_per_period": 100,
      "forecast_period_count": 3,
      "id": "delivery-current",

What you get back

This is the actual output of running the example above — computed by the same function the platform calls, not an illustration.

{
  "configuration": {
    "minimum_history_periods": 12,
    "minimum_loss_observations": 5,
    "random_seed": 53,
    "simulation_count": 1000,
    "tail_probability": 0.05
  },
  "decision": "forecast_supported",
  "exposure_forecasts": [
    {
      "conditional_value_at_risk": 30919.5061,
      "expected_alert_count": 29.406,
      "expected_duplicate_alert_count": 8.521,
      "expected_loss": 7563.1903,
      "expected_missed_material_conditions": 3.305,
      "exposure_id": "delivery-current",
      "failed_gates": [],
      "probability_any_saturated_period": 0.679,
      "supported": true,
      "value_at_risk": 24907.2572
    }
  ],
  "failed_gates": {
    "attention_state_evidence_verified": true,
    "scenario_evidence_verified": true,
    "unsupported_exposure_or_group_ids": []
  },
  "guardrails": [
    "Attention states are aggregate operating conditions, not diagnoses of individuals; do not use them to rank, surveil or discipline named people.",
    "Zero-alert and zero-incident periods, duplicate episodes and mature missed outcomes must be retained or the storm and miss rates are invalid.",
    "The model forecasts under governed scenarios; it does not prove an alert caused an action or a missed alert caused a loss."
  ],
  "method": "markov_modulated_gamma_poisson_alert_fatigue_loss_v1",
  "summary": {
    "conditional_value_at_risk": 51795.5564,
    "expected_duplicate_alert_count": 8.521,
    "expected_missed_material_conditions": 3.305,
    "expected_total_alert_count": 29.406,
    "expected_total_loss": 8833.1903,
    "expected_unique_common_loss": 1270,
    "probability_any_saturated_period": 0.679,
    "value_at_risk": 27711.642
  }

How it works

Markov & state-space control — Choose a policy over evolving states when today's action changes tomorrow's options.

1. 1 Build consecutive zero-inclusive alert-class periods with stable aggregate attention states, condition exposure, unique episodes, total/duplicate alerts, acknowledgement, timely action, interruption and mature missed-loss moments.
2. 2 Update condition frequency, duplicate branching, state-specific acknowledgement/action and state transitions with tenant evidence; simulate coherent workload/capacity scenarios and load-dependent Markov attention transitions.
3. 3 Compound missed material-condition severities and interruption cost, count shared alert-storm loss once, report saturation/miss and VaR/CVaR, and abstain below history, loss or provenance support.

Before you trust it

Every tool in the catalog ships with the conditions under which its answer is meaningful — and the conditions under which it should abstain instead of guessing.

Calibrating it to your org

Calibration workflow

1. 1 Define the management decision, target outcome, aggregate unit, privacy boundary, cadence, and prediction/intervention horizon for this organization.
2. 2 Build a tenant-scoped historical cohort using only information available before each prediction or decision; preserve zero periods, censoring, assignment probabilities, and unresolved outcomes when the method requires them.
3. 3 Estimate statistical parameters on training history, but obtain costs, utilities, risk tolerance, practical-effect thresholds, capacity, and policy constraints from accountable decision owners.
4. 4 Validate on later time windows or held-out aggregate units at the deployment grain, against a simple baseline and the function-specific validation strategy.
5. 5 Deploy only if the returned decision clears evidence, overlap, calibration, robustness, and guardrail checks; warning, unsupported, schema-gap, and fallback decisions are abstentions.
6. 6 Monitor realized outcomes, data drift, coverage, and decision regret; recalibrate at a governed cadence or after a detected regime/definition change, never merely because a stakeholder dislikes the result.

Call it from your AI

You don't wire up 388 tools in your MCP client. The GitRevio MCP server exposes 18 tools, three of which let an agent search the catalog, read a tool's schema, and run it — so the assistant finds this one on its own.

gitrevio_capabilities_search
  { "q": "forecast alert storms duplicate notifications aggregate" }
  → finds "forecast_alert_fatigue_and_missed_risk_loss"

gitrevio_capability_describe
  { "capability_id": "forecast_alert_fatigue_and_missed_risk_loss" }
  → returns the input schema and agent guidance shown on this page

gitrevio_capability_run
  { "capability_id": "forecast_alert_fatigue_and_missed_risk_loss", "arguments": { ... } }
  → returns the result shown above

Works in Claude Desktop, Claude Code, Cursor, Cline, Continue.dev, Goose and Aider. See the MCP server.

Related tools

See every tool in Quality, incidents & reliability →