Past Indeterminacy In Data Warehouse Design

Traditional data warehouse design methods do not fully address some important challenges, particularly temporal ones. Among them past indeterminacy is not handled systematically and uniformly. Furthermore, most methods published until now present transformation approaches by providing examples rather than general and systematic transformation rules. As a result, real-world applications require manual adaptations and implementations. This hinders scalability, long-term maintenance and increases the risk of inconsistency in case of manual implementation. This article extends the Unified Bitemporal Historicization Framework with a set of specifications and a deterministic process that defines simple steps for transforming a non-historical database schema into a historical schema allowing data evolution and traceability, including past and future indeterminacy. The primary aim of this work is to help data warehouse designers to model historicized schema based on a sound theory ensuring a sound temporal semantic, data integrity and query expressiveness.