isop Dataset

National safety investigation organizations operate a continuous lifecycle spanning public oc-currence intake, operations-center triage, multi-source situational context, investigation workflows, evidencehandling, and controlled publication. These functions typically exist as separate portals lacking unified, case-centric back-ends that support cross-channel consolidation, event-time enrichment, auditability, and policy-governed automation. This paper presents N-iSOP, a reproducible reference architecture and benchmarkspecification for multimodal safety investigation operations using the Nigerian Safety Investigation Bureauonline service as a motivating case. N-iSOP retargets an established three-layer functional model (integration,processing\/intelligence, and control\/service) and cloud\u2013edge\u2013device implementation patterns into a bureau-grade \u2018\u2018case graph\u2019\u2019 platform with event-driven microservices, stream processing, and zero-trust securitycontrols. The core extension is agentic orchestration: a policy-bound, tool-using agent that consumes caseevents and generates auditable proposals for deduplication, triage, follow-up tasks, incident-window contextenrichment, and publication metadata. Reliability is enforced through knowledge-grounded prompt chainingand structured outputs, aligned with KG+LLM emergency decision support and disaster management LLMgovernance. We define a benchmark suite with minimal datasets measuring (i) cross-channel entity resolutionquality, (ii) triage and enrichment latency under cloud\u2013edge placement, (iii) authorization\/audit overheadunder Zero Trust, and (iv) custody trace completeness for digital evidence