以基因演算法建立承包廠商最佳化預定進度S曲線

Optimizing Contractor's S-Curve Using Genetic Algorithms

羅維(Wei Lo) ； 陳懿佐(Yih-Tzoo Chen)

《中國土木水利工程學刊》 21卷2期 (2009/06) Pp. 207-217

https://doi.org/10.6652/JoCICHE.200906_21(2).0008

摘要

進度管控與資源規劃是專案管理的核心，幾乎所有的專案管理工作都在進度管控的架構下進行，而其中的專案資源配置（Resource allocation）與整平（Resource leveling）是專案管理中最具挑戰性的工作，亦是決定專案成敗的重要關鍵，良好的資源配置與整平可以使承包廠商進行有效率的運用資源，在最低成本下完成契約工項，達成最大化利潤的目標。本研究考量了過去研究者皆未曾探討的資源互易關係（Resource trade off）及動復員成本等因素對於資源排程之影響，以傳統要徑法（Critical Path Method, CPM）的最早時程計畫為基礎，在總浮時範圍內移動作業開始及結束時間，並利用基因演算法建立求解成本最低之承包廠商最佳化S曲線模型及時程計畫，該最佳化S曲線模型，提供在最晚時程S曲線與最早時程S曲線兩個極端情況所包絡的範圍內，一個明確且成本最佳化的S曲線作為專案執行階段進度管控的基準。為使資源配置模型能夠達到專案整體最佳化更並真實的模擬現實世界的狀況，在資源整平的部份，本研究合理考量專案資源互易的關係並改善傳統一次矩最小的整平方法，同時考慮資源成本之權重，及資源進出專案之動復員成本、閒置成本之條件下，使整平的效益不侷限在單一資源數量，更提升為專案整體資源利用之最佳化，使專案管理者在資源時程規劃階段，即能將資源的調度與運用作有效率的規劃，得以成本最低的方式作資源動員、整合與運用，進而達成追求利潤最大化的目標。

關鍵字

S曲線 ； 基因演算 ； 資源配置 ； 最佳化

並列摘要

Progress control and Resource planning are critical success factors for the project management. While striving to satisfy owners' needs, contractors always intend to minimize construction cost and pursue the maximum profits under a fixed contract amount and duration. And. only a sophisticated work plan, which includes optimal working sequences and perfect timing for executing each individual activity, can optimize the allocation of construction resources and enable contractors to fulfill the contract at the lowest cost. To improve the traditional way of optimizing the resources allocation through resource leveling, this research includes the considerations of costs for resource mobilization/demobilization and idle time and the assumption that for a specific task, there is a trade-off relationship in terms of work productivity between two different types of resources. And, by using Genetic Algorithm (G.A.), this research successfully proposes a contractors' optimal S-curve model, which can be used to identify the near optimal schedule including the optimal quantities for each resources and the timing of performing each activity and ultimately obtain the lowest overall construction cost. Furthermore, by using case study this research demonstrated that an optimal S-curve with the lowest construction costs can be efficiently and effectively identified and used as a baseline to manage the project resources and control the overall project progress.

並列關鍵字

S-curve ； genetic algorithms ； resource allocation ； optimization