資料倉儲是針對決策支援系統的需求所發展出的新一代資料庫的觀念，其資料通常經由線上分析處理，提供管理者決策時的參考。為縮短查詢的時間，並提供使用者各個不同的觀察角度，這些資料通常在某一主題的關聯下，以多維度的資料型式儲存，稱為資料方體。資料方體選取的問題即是，給定一主題及相關的維度所組成的資料方體，考慮使用者欲進行的查詢問題，探討在有限的儲存空間限制下，如何選取適當的子方體（視域）加以實體化，以縮短查詢的時間，這個問題已知是屬於非多項式時間完成問題。因此，目前已知的資料方體實體化的選取方法大多屬於啟發式的方法。本論文旨在比較這些啟發式的挑選方法，分析其效率及求解的品質，以了解這些方法的優劣及適用性。

Data warehousing is a new database concept dedicated to supporting executive managers in decision-making through online analytical processing (OLAP). To decrease the query time and provide various viewpoints, these data usually are organized as a multiple dimensional data model, called data cubes. The data cube selection problem is, given the set of user queries and a storage space constraint, to select a set of materialized sub cubes from the data cubes to minimize the query cost, such as response time and/or the maintenance cost. This problem is known to be a NP-complete problem. Most of the existing algorithms are based on the greedy paradigm. In this paper, we compare and analyze the performance and quality of these greedy selection methods to rank their superiority and suitability.