建築資訊模型(Building Information Modeling, BIM)是現代建築設計的新作法，比起過去以電腦輔助繪圖技術來進行建築設計，BIM除底層技術基於三度空間製圖外，所有建築元件亦可有豐富的屬性資訊，用來表達建築物設計或施工階段的各式資訊需求。當建築物進入完工與營運維護階段時，因建築物外觀與內部結構等主要特徵已確定，對應到BIM軟體內存放的建築物資料也應無異動。從資料庫的角度來看，因目前商用或開放軟體式之建築資訊模型軟體本質上為模型編修用，對於建築物模型與其元件的新增、修改與刪除等動作可謂常用功能，但一旦資料穩定卻有大量查詢需求時，目前BIM技術似無對此最佳化。由於BIM具有3D資料處理能力，本研究嘗試結合BIM與資料庫技術中空間資料處理能力著名的PostgreSQL+PostGIS，以截長補短的方式探討如何應付建築物營運維護階段的新資訊需求。首先將BIM的幾何資訊轉到空間資料庫，並利用資料庫查詢速度快的優勢，在空間資料庫內計算完之後傳回BIM元件ID做後續資訊應用。由於建築資訊模型的幾何計算占大部分的比例，所以利用本研究的方式，可以讓程式執行加快，並讓BIM更適用成為建築物營運維護階段的資訊管理中心。;The building information model (BIM) technology is a new way to design buildings, and every aspect of buildings can be stored in the BIM database. BIM changes not only the traditional design process but the project stakeholders’ responsibilities. Hence, other software applications such as facility management, disaster mitigation, and energy savings should use BIM as the base data representing their buildings and add customized functionalities into BIM so that the goal regarding comprehensive life cycle information management of buildings can be achieved. The spatial database technology is a new way to manage spatial-related data. Compared with other types of databases or technologies such as GIS, a spatial database can process each geometry data type very efficiently and effectively and utilize SQL-based, standardized spatial functions to provide advanced geometry calculation and data processing. Hence, this research aimed at synthesizing BIM and a renowned spatial database, i.e. PostgreSQL with PostGIS, to provide integrated data processing services such as query and attached files manipulation. A data transformation tool from a BIM authoring tool, i.e., Revit, to the spatial database has been developed and it utilizes the bounding box approach to transform each building element. The query performance has been significantly improved by using the synthesized approach and any types of related files can be attached to a set of selected elements. Similar to the index approach used in a typical database application, the proposed approach can provide a new way for facility management users to manipulate BIM and its extension data.
