Temporal Integrity Constraint Management for Sustainable Relational Data

Integrity constraints form a fundamental mechanism for ensuring the correctness, consistency, and semantic validity of relational data. Conventional approaches, including domains, check constraints, assertions, and triggers, provide effective support for static validation but offer limited capabilities for handling evolving business rules and time-dependent integrity requirements. Modern information systems increasingly operate in dynamic environments where constraints may become active, inactive, or modified according to temporal schedules, regulatory changes, or operational policies. This paper addresses the limitations of traditional constraint management by introducing a Temporal Constraint Management Framework (TCMF) that extends integrity enforcement with temporal awareness of integrity constraints. The proposed framework treats constraints as metadata-driven temporal entities associated with validity intervals, activation policies, evaluation strategies, and execution modes. A centralized Temporal Evaluation Engine (TEE) dynamically determines applicable constraints according to valid time, transaction time, and operational context, while a scheduler-based mechanism supports automatic activation, expiration, and reevaluation of temporal rules. The framework enables the coexistence of multiple constraint versions, preserving historical consistency while allowing the introduction of new validation policies without compromising previously valid data. The architecture supports domain constraints, check constraints, assertions, and triggers, as well as custom procedural validations within a unified temporal model. Experimental evaluation conducted on Oracle Database 26ai using a large-scale aviation dataset comprising up to 100 million records demonstrates that the proposed approach reduces processing overhead compared with conventional temporal implementations based on procedural triggers. Results indicate reductions of up to 29.86% in processing costs and 26.88% in execution time while maintaining scalability and distributed synchronization capabilities. The findings confirm that temporal integrity management can be integrated into relational database systems without prohibitive performance penalties, providing a practical foundation for adaptive, context-aware, and evolution-oriented database architectures.