Home Case Studies Ontology Use Cases Decision Patterns Models Assessment About

Predictive modelling systems

Forecast future system states

← Decision Patterns Monitoring & Data

What problem does this solve?

Reacting to events after they happen costs lives and money. Decision-makers need advance warning of floods, droughts, and system failures.

How it works

Computer models combine real-time sensor data with weather forecasts and historical patterns to predict what will happen hours or days ahead, giving time to prepare.

Typical infrastructure

Hydrological models, weather prediction systems, computing infrastructure

Typical monitoring

Forecast accuracy tracking, model calibration against observed events

Strengths

Extends warning lead times; supports proactive rather than reactive decisions; improves with more data over time

Trade-offs

Models are only as good as their input data; extreme events may exceed model assumptions; requires specialist expertise to build and maintain

Related use cases

Operational scenarios where this pattern is applied:

Case studies