Core · What breaks from normal, how likely it is and what changes it
Foresight
Foresight reads what happened to anticipate what comes next: what breaks from normal, how likely it is, what it would hit and which decision changes it.
An engine that projects failures, stock-outs and availability on ontology data, and simulates what happens if a budget, a contract or a capacity changes.
Measuring what happened is not enough to decide. Whoever runs a fleet, a network or a portfolio needs to know what will happen and which lever changes it, before committing the budget.
The engine uses each asset's history —usage, failures, repairs, spare parts, contracts— to project which subsystem will fail, which stock will run out and how much availability there will be. The simulation moves one variable at a time and shows the effect over time, with the assumptions in view.
Today it powers Bastion, including Fleet Control, and the same method serves a mining company, a utility or a police force. In Focus it also projects crime in each municipality five years ahead.
What it delivers
- Failure prediction by subsystem
- Stock-out and reorder point
- Projected availability
- Decision simulation with visible assumptions
What changes before it makes the news
Weak Signals
Every municipality, topic and actor has its normal rhythm. Weak Signals flags when something breaks from it, before it makes the news.
A rise in crime in one municipality, more press coverage of a company, a protest that keeps recurring: each fact on its own looks like noise, and nobody raises a hand.
Weak Signals compares the last week with the previous eight, by municipality, by event type and by actor, and measures how far it departs from what was expected. In each municipality it also compares the current figure with its own history and with the previous year. Only events with sufficient credibility count, and a reading with little history is marked as preliminary.
Every signal comes with its severity, its calculation and its sources in view. The advantage is not seeing more data, but seeing it earlier.
What it delivers
- Last week against the previous eight
- By municipality, event type and actor
- Anomalies against each municipality's own history
- Only events with sufficient credibility
- Severity, calculation and source on every signal
How likely it is and what it would hit
Exposure
For every municipality, institution, company or asset, Exposure calculates how likely something is to happen in the next seven days and what would be exposed if it does.
Knowing a risk exists is not enough to set priorities. Decision-makers ask something else: how likely is it this week, and how much of what is ours — people, assets, money, contracts — is at stake.
On the events in the ontology, Exposure calculates by municipality, topic and actor the probability of at least one event in the next seven days. For each subject it gathers its exposure — population, assets, amounts and contracts —, the candidate risks with their probability and the chain of impact. Base, adverse and severe scenarios are written on that data, without adding figures.
Each organization defines what it watches — an entity or its tax ID, a municipality, a topic — and receives an alert when a warning appears, a trend rises or an event touches what is being watched.
What it delivers
- Probability of at least one event in 7 days
- Exposure: population, assets, amounts and contracts
- Candidate risks and chain of impact
- Base, adverse and severe scenarios on ontology data
- Watches per organization, with alerts
Want to see it with your own data?
We'll show you Serendyp on your country, your sector and your risks.