Schedule Risk Index: What the Numbers Mean

Executives like a single risk number. Schedulers know that number is dangerous unless everyone shares a definition. Bellator’s Schedule Risk Index (SRI) is explicitly a deterministic, pre-simulation indicator on a 0–100 scale—designed to answer: “How fragile is this network if reality wobbles?” not “What is P80 finish?”

Endpoint: POST /api/v1/health/schedule-risk-index (Health family). Simulation is not implemented. Retired POST /api/v1/forensics/schedule-risk-index is a 308 only.

SRI vs health score vs Monte Carlo

Signal	Question it answers	Nature
Health score	Is the model structurally sound and standards-aligned?	Deterministic quality
SRI	How risk-prone is the structure even if “clean”?	Deterministic risk posture
Monte Carlo (Tier 3 planned)	What is the distribution of finish dates given uncertainty models?	Probabilistic simulation
Delay analysis	Who owned historical slip?	Forensic attribution (AACE 29R-03)

Example: a schedule can score Health B (good logic hygiene) and still show elevated SRI because everything sits on a knife-edge critical path with multiple merge points.

How to read the 0–100 scale

Exact banding can be productized in UI copy; use this communication guide:

SRI region	Plain-language posture	Typical management move
Lower third	More resilient structure	Standard monitoring
Middle	Material fragility	Protect float; review merges & long-leads
Upper third	High structural risk	Deep dive before commitment; consider simulation

Higher SRI ⇒ more structural risk. Do not invert it like a “health” score.

Always pair SRI with:

1. Critical path length and density
2. Float distribution (how many days of cushion exist)
3. Constraint load
4. Open-ended or high-lag logic that hides true path length

Drivers you should expect behind the number

Without turning this into an algorithm dump, SRI-class indicators usually worsen when:

• Near-critical mass is high (many tasks with tiny total float)
• Merge bias stacks multiple feeders into one milestone
• Long durations hide internal risk inside a single bar
• Hard constraints create artificial criticality
• Calendars compress effective float
• LOE/summary pollution is not excluded (Bellator excludes classified LOE/summary/hammock from topology stats)

If import classification is wrong, SRI and health both lie. See P6 XER import best practices.

Worked comparison

Two milestone-driven projects, same finish target:

Project	Health	SRI	Story
North Plant	88 (B)	32	Clean logic, healthy float pockets
South Plant	86 (B)	71	Clean logic, but five trades merge into one commissioning gate with ≤2 days float

Leadership should not treat these as equal risk. South Plant needs float protection and maybe a probabilistic study before locking incentives.

Minimal call pattern

curl -X POST "https://api.bellatorsi.com/api/v1/health/schedule-risk-index" \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -d '{
    "tasks": [
      {"id": "A", "duration_days": 10, "predecessors": []},
      {"id": "B", "duration_days": 10, "predecessors": []},
      {"id": "C", "duration_days": 5, "predecessors": ["A", "B"]}
    ],
    "status_date": "2026-07-01"
  }'

Enrich with baselines, constraints, and task_type for production-grade reads. Confirm current request fields in the API reference and playground.

When to escalate beyond SRI

Escalate to deeper tools when:

• SRI is elevated and contractual liquidated damages are material
• Weather, productivity, or third-party uncertainty dominates (needs simulation inputs)
• You are in a dispute (use AACE 29R-03 delay analysis, not SRI alone)
• The model fails DCMA structural checks (fix integrity first)

Tier 3 Monte Carlo and related simulation APIs are planned; SRI is the deterministic bridge available now.

Portfolio use

For PMO dashboards:

1. Import each program schedule nightly.
2. Store health, SRI, and top three findings (not full task dumps).
3. Sort the portfolio by SRI descending within each health band.
4. Trigger human review when SRI jumps week-over-week without scope change.

That pattern catches silent float burn before finish variance explodes.

Next steps

• How to read a CPM schedule health score
• Standards & tier coverage
• Playground — try risk-oriented sample schedules

Sources

• AACE recommended practices on schedule risk and forensic analysis (context for when to simulate vs. attribute)
• GAO Schedule Assessment Guide — valid critical path as a prerequisite to risk claims
• DCMA schedule assessment concepts (float, critical path, baseline execution)
• Bellator standards documentation