Scenario intelligence generates the feasible options for a decision — always including doing nothing — and simulates each one’s outcome, uncertainty and trade-offs before anyone commits, using probabilistic forecasts, optimization and digital twins of the assets and processes involved.
What scenario intelligence does
Always price “do nothing”
The most common decision is the one nobody makes. Scenario intelligence forces an explicit cost on waiting: a compressor that runs to the turnaround, a tank filled to full height, a crack that keeps growing. On this site’s digital twins, the fast-forward control shows exactly that — the cost of acting now compared with the cost of waiting.
Digital twins as scenario engines
A digital twin is a live model of a physical asset or process. Used for decisions, it becomes a scenario engine: fast-forward degradation, test a maintenance window, see readiness change. Explore the plant twin, the storage tank twin and the aircraft readiness twin.
Example: three options, priced
| Option | Cost now | Expected cost later | Downtime | Risk |
|---|---|---|---|---|
| Run to turnaround (do nothing) | $0 | $4.2M × 62% probability | 6 days if it trips | High |
| Swap bearings in planned slowdown | $180K | Low | 36 hours, planned | Low |
| Shut down now for overhaul | $1.1M | Low | 3 days | Low |
Illustrative figures from the plant twin. The option that looks cheapest today — doing nothing — carries the largest expected cost.
Methods behind scenario intelligence
Probabilistic forecasting
Demand, failure and price distributions instead of single numbers.
Optimization
Best feasible plan under constraints such as capacity, crews and contracts.
Simulation
Discrete-event or Monte Carlo models of how a plan unfolds.
Digital twins
Asset-level physics or data models that show condition changing over time.
Show uncertainty honestly
| Practice | Why |
|---|---|
| Use P10 / P50 / P90 ranges | Point forecasts hide risk; ranges reveal it. |
| Show the assumptions | Price, demand, failure rates — decision makers must see what drives the answer. |
| Name what would change the answer | “If Supplier B capacity is below 2,000 units, choose split production.” |
| Compare against realized outcomes | Calibration builds trust in future scenarios. |
- Always include and price ‘do nothing’.
- Ranges beat point forecasts.
- Digital twins turn simulation into something operators can see.
- Tell people what would change the recommendation.