建築資訊模型(Building Information Modeling，BIM)近來廣泛應用，國內外已有諸多文獻指出BIM於工程上的效益，包含工程成本減少、工期縮短、整合工程界面、預先發現問題，另外許多研究也提出分析BIM效益的方法，如KPI法(Key Performance Indicators )、資料包絡分析法(Data Envelopment Analysis，DEA)、ROI法(Return on Investment)及自建BIM模型等方式，然而KPI法權重設定過於主觀，資料包絡分析法要求輸入變數複雜且需十分精確，ROI法僅能分析有明確量化的案例，自建模型法則需耗費諸多時間與人力，無法計算大量BIM案例，此外工程專案具有單一性與無法重複性，將應用BIM與未應用BIM之工程專案進行比較時，其評估基準不同。 有鑒於此，本研究以案例式推理之「最近鄰居迴歸法」建置計算引擎使不同工程專案可於相同或接近相同基準下進行應用BIM之效益比對，提供快速客觀評定BIM應用效益之模式。 In recent years, people understanding of BIM in the architecture, engineering, and con- struction (AEC) industry has improved significantly. BIM can be diversely recognized as a virtual design and construction environment, a communication vehicle among stakeholders, a lifelong information model, or an education platform that can be used for students. A few past research efforts have done to measure the benefits of BIM. However there are no two identical projects, how to evaluate the benefits of BIM between two different projects has been an issue in the previous evaluate approaches. The research adopts the K-Nearest Neighbor Regression of Case Base Reasoning and proposes a model to evaluate the benefits of BIM between different projects to achieve a more accurate evaluation.
