Addition is one of three elementary operations in evaluating well formed (partial) relevance diagrams. An arrow can be added from any node, A, to any other chance node, B, as long as no cycles are created. Adding the arrow replaces B's probability distribution by one that repeats that same distribution across all outcomes of node A. After the arrow has been added, the structure of the influence diagram no longer explicitly represents B's conditional independence with respect to A, given B's direct predecessors. (It remains numerically represented.) The resulting influence diagram is decision equivalent to the original and represents the same joint probability distribution over its uncertain variables.

See also: Linking, Removal, Reversal, Well-Formed Relevance Diagram and Well-Formed Partial Relevance Diagram.