營建專案在長期工程標案競爭下，承包廠商對於工程之預備資金、工料成本及風險評估日益成熟，故競標廠商大多是由具有經驗之承辦人員，以其專業知識、經驗，利用主觀經驗判斷法進行工程成本初期推估。但不同的承辦人員主觀判斷容易造成估價結果有所差異，無法精確地訂定工程底價，且營建工程成本具有眾多不確定性因素，無法讓承辦工程人員有效地評估可能面臨的風險程度。因此，本研究希望利用機率估算來預估工程專案初期成本，彌補傳統估算的不足，首先，本研究將利用AbouRizk(1992)所提出的θ1-θ2座標圖並新增了三項機率分佈函數，結合KNN演算法、留一交叉驗證法、歐幾里得距離，形成一新的機率分佈配適度判定方法以利用在工程項目成本的機率分佈配適度檢定上。最後本研究使用五個公路工程中較常使用到的工程項目進行案例驗證，分別利用K-S檢定法、A-D檢定法和θ1-θ2座標圖進行配適度檢定，對於檢定出來的機率分佈分別繪製其累積機率曲線圖，而由累積機率曲線圖可發現，θ1-θ2座標圖求解出來的機率分佈其累積機率曲線與樣本的累積機率曲線最為接近，也就是說若以θ1-θ2座標圖進行配適度檢定，可求出與樣本配適度較高的機率分佈函數。藉由求得的機率分佈函數，可再進一步推估其成本合理估價範圍。而對於營造工程成本的預測已有許多學者提出相關的研究和探討之成果，因此本研究是希望藉由提出不同的研究方式能夠提供業界學者有著不同的參考應用。

Construction engineering projects in the long-term competitive tenders, contract manufacturers for prepare funding, quantity cost and risk assessment of the project has become more sophisticated. Therefore, most of the companies bidding by the contractor personnel with experience, with its expertise, experience, initial engineering costs reckoning use of subjective experience to judge by. But the difference is likely to cause the contractor personnel subjective valuation results to differ, not accurately set the reserve price engineering, and construction cost of the project has a number of uncertainties, can not let the contractor engineering personnel to effectively assess the degree of risk that may be encountered. Therefore, this study hopes to use to probability estimate to estimated project initial cost, compensate for the lack of traditional estimates. First, the present study will applied AbouRizk (1992) proposed θ1-θ2 plots and added three probability distribution function, Combined KNN algorithm, Leave One Out Cross Validation method, the Euclidean distance, the formation of a new method for determining the probability distribution of fit in to take advantage of the probability distribution of project cost with Goodness of fit test. Finally, this study used five more commonly used in highway engineering projects to be verified cases were the use of Kolmogorov-Smirnov Test, Anderson-Darling Test and θ1-θ2 plots of fit test. For probability distributions are plotted identified Cumulative Distribution Function graph, and the cumulative probability graph can be found, the chances of solving θ1-θ2 plots the cumulative probability distribution curve out of the cumulative probability curve with the sample closest, that terms of θ1-θ2 coordinate diagram of fit test, can be obtained and samples with moderately high probability distribution function. Determined by probability distribution function can further reckoning the cost reasonable valuation range. As for the cost of the project to create a prediction has been made by many scholars and research achievements related to the discussion, so this study is to be presented in different ways to provide industry research scholars have different reference applications.

參考文獻
1.AbouRizk, S. M., and Halpin, D. W., “Probabilistic simulation studies for repetitive construction processes,” Journal of Construction Engineering and Management, ASCE, 116(4), pp.575~594, 1990.
2.AbouRizk, S. M., and Halpin, D. W., “Statistical properties of construction duration data,” Journal of Construction Engineering and Management, ASCE, Vol.118, No.3, pp. 525-543, 1992.
3.Back, W. E., Boles, W. W. and Fry, G. T. “Defining triangular probability distributions from historical cost data.” ASCE Journal of Construction Engineering and Management. 126(1), 29-37. January/February, 2000.
4.Cavalieri, S., P. Maccarrone, and R. Pinto, Parametric vs. neural network models for the estimation of production costs: A case study in the automotive industry. International Journal of Production Economics. pp. 165-177, 2004.
5.Cheng Yeh, “Quantity Estimating of Building with Logarithm-Neuron Networks” Journal of Construction Engineering and Management, Vol. 124, No. 5, pp. 374-380, 1998.
6.Chou, J.-S., et al., Quantity-based approach to preliminary cost estimates for highway projects. Transportation Research Record , pp. 22-30, 2006.
7.Cristiano Maio, Cliff Schexnayder, Kraig Knutson, Sandra Weber, “Probability Distribution Functions for Construction Simulation”Journal of Construction Engineering and Management,ASCE, Vol.126,No.4,pp. 285-292, 2000.
8.Crowley,L.G., Hancher,D.E., “Evaluation of Competitive Bids.” Journal of Construction Engineering and management, ASCE, Vol.121, No.2, pp.238-245, 1995.
9.Dozzi, S. P., AbouRizk, S. M., and Schroeder, S. L. “Utility-theory model for bid markup decisions.” ASCE Journal of Construction Engineering and Management. 122(2), 119-124, 1996.
10.Elkjaer, M., “Stochastic Budget Simulation.” Journal of Project Management, Vol.18, pp.139-147, 2000.
11.Hojjat Adeli, and Mingyang Wu, “Regularization Neural Network for Construction Cost Estimation”, Journal of Construction Engineering and Management, Vol. 124, No. 1, pp. 18-24, 1998.
12.Kamal M. Al-Harbi, David W. Johnston, and Habib Fayadh, “Building Construction Detailed Estimating Practices in Saudi Arabia”, Journal of Construction Engineering and Management, Vol. 120, No. 4, pp. 774-784,1994.
13.Kitchenham, B.A., et al., A framework for evaluating a software bidding model. Information and Software Technology, pp. 747-760, 2005.
14.Niazi, A., et al., Product cost estimation: Technique classification and methodology review. Journal of Manufacturing Science and Engineering, Transactions of the ASME, pp. 563-575, 2006.
15.Ntuen, C. A., & Mallik, A. K.“Cost Estimation As A Complexity Modeling Problem. ”In AACE International. Transactions of the Annual Meeting. AACE, 1987.
16.Oberlender, G.D. and S.M. Trost, Predicting accuracy of early cost estimates based on estimate quality. Journal of Construction Engineering and Management ,pp. 173-182 , 2001.
17.Rafael Romeu, “An Intraday Pricing Model of Foreign Exchange Markets”IMF working paper, 2006.
18.Schexnayder, C., Knutson, K., and Fente, J. "Describing a Beta Probability Distribution Function for Construction Simulation." J. Constr. Eng. Manage., 10.1061.ASCE, pp.221-229, (2005).
19.Skitmore, M.,Drew,D. and Ngai,Stephen, “ Bid-Spread.” Journal of Construction Engineering and management, ASCE, Vol.127, No.2, pp.149-153, 2001.
20.Tarek Hegazy, and Amr Ayed,” Neural Network Model for Parametric Cost Estimation of Highway Projects” ,Journal of Construction Engineering and Management, Vol. 124, No. 3, pp. 210-218, 1998.
21.Thomas M. Walski, ”Estimating O&M Cost when Costs Vary with Flow”, Journal of Water Resources Planning and Management, Vol. 110, No. 3, pp. 355-360, 1984.
22.Touran, A., and Wiser, E., “Monte Carlo technique with correlated random variables.” Journal of Construction Engineering and Management, ASCE, Vol.118, No.2, pp.258-272, 1992.
23.Touran, A., “Probabilistic Cost Estimating with Subjective Correlations.”Journal of Construction Engineering and Management, ASCE, Vol.119, No.1, pp.58-71, 1993.
24.Uher, T.E., A Probabilistic cost estimating model. Cost Engineering (Morgantown, West Virginia), pp. 33-40, 1996.
25.Yang, I.T., Simulation-based estimation for correlated cost elements. International Journal of Project Management, pp. 275-282, 2005.
26.Wall, D.M., Distributions and correlations in monte-carlo simulation. Construction Management and Economics. p. 241-258, 1997.
27.W. Edward Back, Walter W. Boles, and Gary T. Fry,” Defining Triangular Probability Distributions from Historical Cost Data”, Journal of Construction Engineering and Management, Vol. 126, No. 1, pp. 29-37, 2000.
28.蘇明通，「政府採購之標價分析」，主計月報，第526期，第44-50頁，1999。
29.李建雄，「建築估價(修訂版)」，詹式書局，1995。
30.王維志、李建中，「成本模擬及單價比對以利採次低標價決標」，第一屆營建管理學術研討會論文集，台北，第三冊，第 105-114 頁，1999。
31.王維志、李建中，「電腦模擬與單價比對在工程採購招標之運用」，中國土木水利工程學會會刊，第二十七卷，第四期，第 3-12 頁，2001.2。
32.劉福勳、鄧朝元，「電腦估價概說」，現代營建雜誌社，p.37-41，1993。
33.陳冠霖，「公共工程經費估價電腦化與營建物價關係」，營建物價，p.5-8，1997。
34.顏月珠，「商用統計學」，三民書局，1993。
35.吳韻青，「營造工程估算機制系統研究」，國立成功大學建築研究所碩士論文，2013。
36.余家祥，「以案例式推理建構建築工程成本估算系統」，國立中央大學土木工程學系碩士論文，2001。
37.韋桂仁，「公共工程標價偏低時之審標作業研究」，國立中央大學土木工程學系碩士論文，2002。
38.許景翔，「廠商投標價決定模式-由分項工程角度切入」，國立交通大學土木工程學系碩士論文，2006。
39.呂翊民，「公共工程技術資料庫整合計畫整合性之探討」，國立交通大學土木工程系碩士論文，2003。
40.謝文山，「演算式建築工程成本概算之研究」，國立台灣科技大學碩士論文，2001。
41.張鈺輝，「距離夾角最似鄰近結合貝氏定理預測分類推論模式之研究」，國立台灣科技大學營建工程系碩士論文，2015。
42.周瑞生，「隨機模擬於項目制專案成本信度區間推求與風險評估研究成果報告」，國立台灣科技大學營建工程系，2010。
43.吳太隆，「機率模型應用在工程底價訂定之研究」，國立台灣科技大學營建工程系碩士論文，2004。
44.張梅英，「建立土地大量估價方法之研究」，國立政治大學地政研究所博士論文，1992。
45.徐碧君，「常態K-S檢定統計量之近似極限分佈函數」，靜宜大學應用數學系碩士論文，2009。
46.曾郁婷，「小樣本的常態性檢定之探討」，真理大學數理科學研究所碩士論文，2008。
47.王蒙納，「營造工程物價指數技術指標預測模式之研究」，正修科技大學營建工程研究所碩士論文，2010。