Capability-Centric Enterprise Architecture for Business-Technology Alignment and AI Adoption

This paper presents an integrated overview of four complementary enterprise architectures

artifacts: Business Capability Maps, Technical Capability Maps, shared technical implementation models,

and business-to-technical traceability matrices. Business Capability Maps capture the stable 'what' of the

enterprise, while Technical Capability Maps capture reusable technology-side abilities independent of

specific products and projects. Shared technical implementation models define how common capabilities

are delivered across teams using shared, federated, or local patterns, improving reuse, governance, and

speed. Traceability matrices provide the relationship layer that links business priorities and capabilities to

technical capabilities, platforms, applications, and controls for prioritization, gap analysis, and impact analysis,

and governance. The paper then explains how to adopt artificial intelligence (AI) across these artifacts by

modeling AI as business value (AI-enabled sub-capabilities), technical enablement (inference, guardrails,

monitoring, auditability), shared platform capability (federated enterprise AI), and DevOps extensions

(MLOps and LLMOps). The proposed approach enables organizations to align AI investments with business

outcomes, risk controls, and operational resilience.