Prospective Completion Prediction for Independent-Trial Cluster Search with Compositional State: Measurement Burden, a Pre-Outcome Separation Gate, and Structural Limits

Abstract Completion prediction for independent-trial cluster search requires productive capacity and worker activation, and the question is acquisition cost as well as accuracy. Under an ideal-hash model, we combine independently characterized state with a known per-trial success probability in a zero-fit analytic predictor. For the compositional representations in two measured configuration regimes, total-completion-time cell-mean absolute percentage errors were 1.56% and 2.54–2.67%, with no accuracy ordering established. For a nine-target acquisition geometry, node-total composition reduced total characterization burden by 35.52%. An unseen-family campaign remains descriptive after a post-outcome corrective. In a fresh characterization-only campaign, predicted occupancy-aware/equalized separation was 0.164%: it passed the prespecified characterization-uncertainty relative-precision heuristic but remained far below the 5% experimental-design continuation threshold, so the frozen protocol authorized no target collection. Structural analysis identifies rate–activation interaction and full-path invariance conditions; an outcome-free diagnostic additionally reveals material occupancy-four within-node worker-rate heterogeneity, limiting physical interpretation of equal-share worker rates.