隨著建築資訊模型（Building Information Modeling， BIM）技術於營建 產業之廣泛應用，BIM 已逐漸成為整合設計與施工資訊之核心媒介。然實 務上仍缺乏標準化之模型品質檢核機制，致使圖說與模型不一致之情事頻 傳。本研究旨在建立一套具實務應用性之「BIM 建築結構自主檢查表」， 藉以提升 RC 大樓設計階段之模型建置之精確度與品質。 本研究採質性與量化並行之混合研究法。首先透過文獻回顧建構初步 檢核架構，繼以專家訪談與問卷調查進行實證分析。專家訪談結果指出， 業界專家一致認同自主檢查表於設計階段具備「一級品管」之預警功能， 能有效降低後續介面整合之衝突。問卷統計分析結果顯示，受測之產學界 專業人士對於「建築構造設計錯誤導致工期延宕及變更設計」之風險認知 具備高度共識；經統計驗證，受測者對於自主檢查表能提升模型幾何資訊 完整性及降低構件定位偏差之認同度均值均達顯著水準。 本研究進一步以實際住商大樓案例進行導入驗證。案例分析結果證 實，透過自主檢查表之應用，能精確偵測出設計瑕疵。例如：發現陽台欄 杆金屬格柵組件於大樣圖與平立剖面圖間存在顯著幾何矛盾，且格柵高程 配置亦有不一之情形。經由檢核機制之回饋與模型修正，顯著提升了模型 與施工圖說之合致性。 本研究所建置之自主檢查表希望能作為建模人員之作業參考，更經由 統計實證與案例驗證確認其於品質管理上效益明顯，應有實務應用價值與 技術參考意義。;With the widespread application of Building Information Modeling (BIM) technology in the construction industry， BIM has become a core medium for integrating design and construction information. However， the lack of standardized quality inspection mechanisms in practice has led to frequent inconsistencies between design drawings and models. This research aims to establish a practical "BIM Architectural and Structural Self-Checklist" to enhance the accuracy and quality of BIM model information for reinforced concrete (RC) buildings. This study adopts a mixed-methods research approach， combining qualitative and quantitative analyses. First， a preliminary inspection framework was developed through a literature review， followed by empirical analyses via expert interviews and questionnaire surveys. The results of the expert interviews indicate that industry experts unanimously agree on the necessity of implementing a self-checklist during the design phase as a "primary quality control" early warning function， which effectively reduces subsequent interface integration conflicts. The statistical analysis of the questionnaires shows a high degree of consensus among academic and industry professionals regarding the risk of construction delays and design changes caused by architectural design errors. Furthermore， statistical verification confirms that the respondents′ agreement on the checklist′s ability to improve geometric information integrity and reduce component positioning deviations was statistically significant. This research further validates the proposed checklist through a case study of a mixed-use building project. The case analysis results demonstrate that the application of the self-checklist can precisely detect design flaws. For instance， significant geometric contradictions were identified between the detail drawings and the plan， elevation， and section views for metal grating components of balcony railings. Inconsistencies in the elevation configuration of gratings were I also observed. Through the feedback mechanism and subsequent model revisions， the consistency between the BIM model and construction drawings was significantly improved. In conclusion， the checklist developed in this study serves as a robust operational reference for modeling personnel. Validated through statistical evidence and case verification， it demonstrates clear benefits for quality management and offers substantial value for practical application and technical reference.
