From Queue Measures to Process Diagnosis

This preprint develops a scope-bound recoverability framework for moving from queue measures to process diagnosis in institutional systems. Waiting times, queue stocks, and backlogs describe unresolved burden but do not by themselves identify the process that generated it. The framework therefore specifies which distinctions must remain recoverable before an institutional queue can support localisation, comparison, or mechanism-specific claims. It separates five task-relative process roles: unresolved state, admitted inflow, realised processing, internal transition-retention, and typed terminal outflow. It also distinguishes case-level outcomes from aggregate regimes, separates nominal resources from effective capacity, and blocks recurrent substitutions such as activity for progression, exit for substantive resolution, and nominal resources for operative capacity. The framework is intended as a diagnostic architecture rather than a universal queue model and yields where an established model represents the required distinctions more clearly or economically. This is Paper II in the research programme Fragmentation, Capacity, and Queue Formation in Complex Institutional Systems.