Decision matrix · human to global · milliseconds to a century
Built Environment Decision Matrix
Decisions in the built environment, organised by spatial scale and time horizon. Each row says who decides, how much can be automated, what a world model would have to predict, and whether that prediction can be simulated. Click a cell in the overview to filter, and click a row for its full details.
Overview: scale × horizon
Number of decisions per cell; the letters show each decision's level of autonomy, and the bar shows how many are simulatable (yes, partly, no). Click a cell to filter the table and draw its relations: teal arrows show what its decisions affect, amber arrows what affects them. Click it again to clear.
Simulatable: yesPartlyNoAutonomy: Aautomatic Ssupervised Ddecision support Hhuman or policy
Cross-scale couplings
How decisions constrain, feed or trigger each other across scales and horizons. Click an ID to show that decision.
Legend
Simulatable
- Yes
- The quantity to predict follows from physics or engineering models (heat, light, flow, traffic) with available data.
- Partly
- Part of the outcome can be simulated, but results hinge on behaviour, markets or occupancy that must be assumed or learned.
- No
- Driven mainly by human preference, organisation, politics or markets: learn from data or use scenarios.
Autonomy
- Automatic
- A controller acts alone.
- Supervised
- Automation acts; a person can override.
- Decision support
- AI recommends; a person decides.
- Human / policy
- People or institutions decide, often collectively.