Enterprise analytics initiatives frequently assume that relevant data can be centralized, standardized, and made available for decision support through a unified platform. In practice, operational decisions often depend on fragmented data distributed across legacy enterprise resource planning (ERP), planning, finance, supplier, logistics, and operational systems. This paper proposes an Interoperability-Driven Analytics Framework to improve decision support without requiring the wholesale replacement of existing systems of record. The framework integrates five components: system-of-record preservation, cross-system signal extraction, semantic alignment, decision-rule orchestration, and traceable decision support. The proposed architecture treats interoperability as an analytics control layer that preserves operational ownership while enabling governed signal extraction, physical data element (PDE) mapping, rule execution, provenance capture, and workflow activation. A discrete-event simulation was conducted over 180 operating days using 14 heterogeneous legacy systems, 320 supplier entities, 42,000 daily operational records, and five architectural baselines: manual reconciliation, point-to-point extract-transform-load (ETL), centralized data lake, API-only integration, and full system replacement. Results show that the proposed framework reduced decision-readiness latency from 31.4 h to 7.6 h relative to manual reconciliation, reduced semantic conflict rate from 18.6% to 4.3%, improved provenance completeness from 42.1% to 93.4%, and increased exception decision coverage from 51.8% to 87.9%. The economic model produced USD 3.47 million in annualized net economic value, with implementation costs 54.8% lower than those of full replacement. The findings position interoperability as a measurable analytics design capability rather than a secondary integration concern.
Thatikonda et al. (Sat,) studied this question.