From Technological Capability to Disaster Governance in North Korea: An Evidence-Bounded Framework from Scientific Publications

Abstract

This study examines the extent to which scientific publications can support claims about algorithmic disaster governance in an information-limited setting. The empirical corpus consists of two purposively selected 2024 North Korean journal issues—Computer and Software Technology and Geologic and Geographic Science. All 48 scholarly or technical items listed in the two issues were screened, and 14 were retained for structured analysis. Using qualitative document analysis and evidence-bounded coding, the study distinguishes among technological capability, functional compatibility, administrative embedding, and system integration. Three inference rules constrain the escalation of claims: greater functional distance weakens evidentiary support; claims of embedding require connective institutional evidence; and module-level capabilities are non-additive for platform-level claims. The sampled literature documents disaster-risk assessment, remote sensing, terrain processing, deep-neural-network water-depth estimation, routing, sensing, robotics, optimization, and participatory-data workflows. Two qualitative robustness tests clarify the limits of inference. Excluding all transfer-hypothesis cases preserves the modular technical pattern; omitting the principal governance source, however, removes direct support for a full-cycle administrative architecture while leaving the technical-capability portfolio and inference framework intact. The article’s general contribution is therefore an evidence-bounded capability–compatibility–embedding protocol. Modular algorithmic disaster governance is the empirical pattern identified by applying that protocol to this bounded corpus, not a claim that an integrated nationwide system has been deployed.