Governing the Recovery of Stalled Systems-of-Systems: ADAPT, a Dependency-Anchored Methodology, with Multi-Case Feasibility Illustrations

Systems-of-systems (SoS) integrate constituents that retain operational and managerial independence, and that independence makes their engineering artifacts prone to drift: a change in one constituent does not propagate, misalignment surfaces late, and outputs accumulate that trace to no validated need. A recurring consequence is the stalled SoS, years of effort across independent teams yielding no usable, connected baseline. The SoS-engineering and complex-system-governance literatures describe what governance should achieve but offer little procedure for recovering an SoS that has lost coherence. This paper introduces ADAPT (Anchor, Dependency, Allocation, Production, Traceability), whose five components operationalize the integrating purposes (communication, coordination, control, and integration) for SoS recovery, enhancing rather than replacing control boards and program offices. It is illustrated through four de-identified cases (feasibility evidence, not a test) under an explicit case-study protocol; in the lead case, six prior teams, comparably resourced, had produced no usable baseline and ADAPT produced an approved one in four months against a documented eighteen-month plan of record. The evidence is retrospective, single-organization, and uncontrolled, supporting the propositions analytically, not statistically; a pre-registered confirmatory study with falsification conditions is specified, and ADAPT’s measures are additionally exercised on an independent public SoS dataset (Dronology/SAFA) as external construct validation.