學校重建工程預算與合約工期之預測

Predicting the Budget and Duration of School Reconstruction Projects

陳維東(Wei-Tong Chen) ； 黃盈樺(Ying-Hua Huang) ； 陳鴻隆(Hong-Long Chen) ； 廖深利(Shen-Li Liao)

《技術學刊》 21卷1期 (2006/03) Pp. 1-18

https://doi.org/10.29507/JT.200603.0001

摘要

儘管營建工程早期精確估算的重要性不言而喻，預算及工期之推估卻少有定量的方法可茲應用，一般多半靠承辦人員本身之專業知識及先前相關工程與工料分析而得，其中主觀成份與不確定性顯著。因此，如何有效地推估合理的預算與工期向來是營建工程的一個重要課題。本研究蒐集彙整921學校重建工程以最有利標及最低標的實際案例，應用統計迴歸方法建構重建工程預算及工期之預測估算模式，以供新進人員、業主或承包商建立規劃階段營建工程成本整體概念。研究結果顯示，本研究所建構之預測模式因使用的變數單純且取得容易，具備相當的方便性；預測模式亦可簡化估算之作業程序，減少對工程經驗豐富人員的依賴；最有利標案例預算金額及合約工期預測模式的平均誤差分別為7.77%與7.67%，最低標案例預算金額及合約工期預測模式之平均誤差則分別為17.17%與15.75%；最有利標似乎比最低標更能縮減工程工期，而最低標似乎比最有利標更能節省成本，惟此發現仍待蒐集更多案例進一步研究探討確認。

關鍵字

工程預算 ； 合約工期 ； 學校重建工程 ； 迴歸分析

並列摘要

There is consensus among researchers and practitioners concerning the importance of accurate estimates during the early stages of construction projects. However, only a limited number of quantitative models and approaches exist that are capable of objectively evaluating the accuracy of early estimates. This lack of models forces the evaluation of early estimates to rely heavily on the professional knowledge of estimators and cost data from previous projects. Thus, development of methods for effectively predicting reasonable budgets and project durations has become a critical issue for the construction industry. This study presents regression models capable of forecasting both project budgets and contract durations for school reconstruction projects. Those projects examined to develop the proposed models were awarded by using two types of tendering system: the low bid system (LBS); and, the most advantageous system (MAS). The proposed models can establish overall cost concepts during project planning phases for owners and contractors. The proposed models are applied easily since their independent variables are easy to obtain. The proposed models can simplify the estimate procedure and, thereby, decrease dependency upon experienced personnel. The average error rates of the prediction models for budget and contract duration for MAS were 7.77% and 7.76%, respectively. The average error rates of the prediction models for budget and contract duration for LBS were 17.17% and 15.75%, respectively. It seems that the MAS was found to be better than the LBS in terms of project duration reduction, whereas the LBS performed better than the MAS in terms of project cost saving. However, more in-depth research regarding cost/duration reduction needs to be carried out to identify the effectiveness of the MAS and the LBS.

並列關鍵字

construction budget ； contract duration ； school reconstruction ； regression analysis