現行之營建工程中常包含甚多之重複性作業,而其排程及資源配置方法對施工規劃人員常成為重大挑戰,所以目前有甚多研究者進行重複性專案排程之研究。近來學者提出工作群組式重複性專案的形式,可以較傳統之重複性專案有更廣泛之應用範圍,因為此種形式能夠表達作業項目間更為一般性的邏輯關係及包括更多樣化之資源使用方式,因此更為符合施工實務之需要。本研究以基因演算法為基礎,提出一個考慮資源數量限制、資源多樣性、以及資源使用連續性等條件限制下之工作群組式重複性專案排程最佳化尋優模式,並以整體專案淨現值最大化為目標。本模式可以為工作群組式重複性專案提供資源指派及作業順序安排之最佳或接近最佳解。經由實際案例測試結果顯示,本研究所提出之最佳化模式可以正確求解工作群組式重複性專案排程,並具有良好的演算效率。
Repetitive activities are commonly found in construction projects. The scheduling and resource allocation of repetitive activities often present great challenge to the project planner. Much interest has been recently devoted researchers' attention to the scheduling of the repetitive project. A workgroup-based project scheme is issued to enlarge the application field of the traditional repetitive project because of it has been able to involve more generalized logic relationship between activities, more diversity of resource used, and more practical for real site condition to meet. An optimal scheduling model for workgroup-based repetitive project with the consideration of the impact of limited quantity of resources, diversity of resource types, and resource continuity is proposed in this paper. A genetic algorithm-based searching technique is adopted to establish the optimal scheduling model. The proposed system can effectively provide the optimal or near-optimal combination of project durations, resource selection/assignment, and activities alignment with the objective of maximum the net present value (NPV) of whole project. The optimization results of the example are very promising and demonstrate the application value of the model.