| 摘要: | 重複性工程是由數個作業內容相同或相似的單元所組成的專案，例如：高樓建築、公路工程、捷運系統、高速鐵路、機場跑道、管涵工程等。對於重複性工程來說，最重要的就是維持工作連續性。工作連續之目的為減少工作團隊在單元間的閒置時間，即能有效降低直接成本、提高生產效率；但為求工作連續，而減少單元間的閒置時間，若是遇到突如其來的狀況，將使得施工團隊一旦在前一單元延遲，將衝擊下一單元的開始時間，進而造成總工期的延誤。
在重複性工程排程時，有效利用作業間的浮時，維持一定的工作連續性，加入施作單元間的緩衝時間，降低施工的閒置時間，使專案成本降低。本研究目的為控制降低重複性工程不確定性風險所帶來的影響，將緩衝時間應用於作業間，以吸收突發事件對於總工期的衝擊，並考量不確定性工期與成本，依據各施工團隊的工作連續狀況，找出時間成本權衡。研究方法則是以蒙地卡羅電腦模擬結合基因演算法之尋優技術，決策目標為一定工期要求內特定信心水準下之最小總成本。應用本研究的解析模式進行實務案例模擬，決定施作單元間之最佳緩衝時間，並與文獻上的案例結果比較及分析。
The repetitive project is the project composed of several units with the same or similar operating contents such as the skyscraper building, the highway engineering, the MRT system, the high-speed railway, the airport runway and the pipe culvert engineering. For the repetitive project, one of the most important issues is to maintain work continuity. The purpose of work continuity is to reduce the idle time among units for the working team, which can efficiently reduce the idle cost and to enhance production efficiency. However, if any unexpected situation occurs, the reduction of idle time among units to meet the demand for the work continuity would cause that once the working team delay in the former unit, the beginning time of next unit will be crashed and then the total working period will be impeded.
When the repetitive project is scheduled, one could efficiently utilize the float time of activities and maintain suitable work continuity; then the buffer time should be imposed among the units and the idle time of working should be reduced to lower the cost of project. The purpose of this research aims to study the impact from reducing the risk of uncertainty in the repetitive project, to apply the buffer time among the operation to dilute the dash of total working period from the unexpected incidents, as well as to consider the uncertain working period and cost and according to the situations of work continuity among each working team, to seek the optimal tradeoff between time and cost. Moreover, this study uses the Monte Carlo Simulation combined with the Optimization of Genetic Algorithms. Goal of decision is to optimize the total cost under the specific confidence level of a particular working duration. Finally, the analytic model of this research will be applied to conduct the practical case simulation, the best buffer time imposed among the units will be decided, and the results will be compared with the case findings from the documents. |
