The correct answer is Data models. DAMA-DMBOK2 explicitly explains that data models and other Data Architecture artifacts are commonly created or captured within development projects and are subsequently standardized and managed by Data Architects.

Development projects routinely produce conceptual, logical, and physical representations of the data required by a solution. If each project develops those structures independently without architectural review, inconsistent terminology, duplicated entities, conflicting definitions, and incompatible relationship structures can accumulate across the enterprise. Data Architects therefore review project models, reconcile them with enterprise standards, and identify structures suitable for reuse.

An Enterprise Data Model is itself an important architectural artifact, but the wording of the question refers specifically to artifacts typically generated in individual development projects and later standardized. The DMBOK2 passage uses data models in precisely that context.

This process has a direct Data Quality benefit. Standardized models improve consistency of definitions, domains, keys, relationships, and constraints. They also provide metadata from which integrity, uniqueness, validity, and completeness rules can be derived.

Reference Topics: DAMA-DMBOK2 Chapter 4 — Data Architecture; Architectural Artifacts; Development Projects; Chapter 5 — Data Modeling and Design; Chapter 13 — Consistency and Integrity.

===============