Cross-Impact Balance Analysis

Classical cross-impact analysis, going back to Gordon and Hayward, works with event probabilities and conditional adjustments, which proved mathematically awkward and sensitive to estimation error. Weimer-Jehle's CIB instead uses ordinal judgments of how strongly one descriptor state promotes or restricts another, and defines scenarios as fixed points of a deterministic balance algorithm. The output is a set of internally consistent configurations rather than revised probabilities, which makes CIB better suited to qualitative, categorical foresight problems.

A scenario is internally consistent when no descriptor would change its state given what all the other descriptors are doing. Formally, for each descriptor the selected state must have the highest impact balance — the largest net sum of promoting minus restricting influences from the rest of the configuration. This is a self-reproducing fixed point: the configuration regenerates itself under the balance rule, so there is no internal contradiction pulling any single assumption in a different direction.

No. CIB filters for internal coherence, not empirical probability. A consistent scenario is one whose assumptions reinforce rather than undermine each other, but coherence and likelihood are different things; an internally consistent future can still be unlikely. Analysts sometimes layer additional information, such as expert plausibility ratings or the number and depth of supporting impacts, to discuss relative plausibility, but the core method deliberately separates consistency from forecasting.