應用電腦模擬技術處理營建專案不確定性排程已獲得營建管理領域廣泛的接受。常見的模擬模式包含下列步驟：(1)以機率分佈為基礎建立模擬元件之相關資訊(如作業需時，機具產能，工作數量等)、(2)轉換為電腦語言、(3)進行多次運算，並獲取特定管理資訊(如生產效率，循環時間，工程總需時等)。 先前有許多學者針對模擬中元件參數彼此存在之統計相關性做相關研究，但大都假設元件參數之機率分佈為單一型式，因此本研究假設參數為多種分佈形式。研究方法為應用多變量模擬中之NORTA（NORmal To Anything）法產生具共變異性之隨機亂數並以此進行模擬運算。應用案例為路面刨除加鋪工程，研究成果為建構模擬模式進行工程總需時之不確定性分析，並協助排程工程師評估作業參數共變異條件對總需時產生之影響。 本研究驗證過程包含兩部份，其一是對於由NORTA法所產生之變數，利用統計檢定驗證是否符合當初假設之分佈型式，另一部份則是檢定產生出之相關係數矩陣，是否接近原來設定之相關係數矩陣。 Using the computer simulation technology to treat the uncertainty in the scheduling of projects has already been accepted by the field of construction management. General simulation models include the following steps:(1)establish the probability distributions of simulation components.(ex: activities duration, production efficiency of machines and tools, the amount of activity.)(2)transform into computer code.(3)perform a number of simulation iterations and obtain specific management information.(ex: production efficiency, duration of cycle, total duration of project.) Previous researches attempted to address possible correlations among the parameters of simulation components. They, however, often assumed that the distributions are of the same form. This study allows different kinds of distributions, and generates correlated random variables by applying the NORmal To Anything (NORTA) method. The proposed model has been applied to a pavement rehabilitation project. This study develops a simulation model to analyze the uncertainty of project duration at the presence of correlations among simulation components. The validation of this study involves two parts. The first part is to test whether the generated samples have the specified marginal distributions. The second is to check whether the generated correlations are close to the specified ones.
