運用專案完成機率與PSO演算法建立趕工專案期望利潤模式之研究

Establishing the Expected Profit Model of the Crashing Project with Project Completion Probability and PSO Method

陳育群(Yu-Chiun Chen)

指導教授 ： 任恒毅

元智大學/工程學院/工業工程與管理學系/碩士(2008年)

https://doi.org/10.6838/YZU.2008.00166

摘要

專案存在許多的不確定性，也是造成專案時間延誤、專案執行所耗的成本超過預算的主要原因。在實務中，專案執行者與專案成本評估執行者常利用自身以往執行專案的經驗進行評估。評估過程時，往往不易在時間與成本之間取得適當權衡，尤其是專案處於進度時程急迫點或是因客戶特別要求而成立的趕工契約，此類情況在營建工程專案中更為明顯。

過去學者研究關於專案作業時間成本權衡分析，多考量求取最佳時間成本權衡之最佳組合，或有考量到作業工期與成本的隨機性，較少分析所求得之最佳解之可行程度，亦即雖有專案時間與成本權衡的最佳解，但卻不易按照此計算之最佳時程或預算時程內完成專案。

本研究旨在估算專案完成的機率與建立專案期望利潤評估之ㄧ般模型，包含：(1)求專案最小直接成本、(2)估算專案完成機率、(3)建立專案利潤評估模型、(4)風險回應策略等四個階段。首先，本模型利用PSO啟發式演算法求取此專案的最小直接成本，再藉由蒙地卡羅模擬方法估算並建立專案完成機率矩陣，據此驗證PSO所求之最小直接成本的完成機率。經由對下水道專案之實證分析結果發現：在建立專案利潤評估模型時，若其完成機率過小，應提高投入成本以提升專案完成機率，並以期望利潤多寡決定專案之最適投入成本與時間，在面臨業主變更並壓縮專案期程之情況，可運用本模型考量以執行此專案，避免專案因延誤造成的風險與相關損失。

關鍵字

時間成本權衡 ； PSO演算法 ； 專案完成機率 ； 專案利潤評估模型

並列摘要

The project uncertainty is always the reason of project delay or budget overrun. In practice, project managers often estimate the project cost and schedule depending on their historical experiences. In the process of estimating both project cost and schedule, it is usually hard to achieve good balance between project duration and cost. Project crashing of construction project, especially makes it hard to get this balance.
Past research of project activity time-cost trade-off mainly focus on the deterministic activity time and costs, with some studies on stochastic time and cost factors to find the solution. But they seldom consider project completion risk, namely the chance to finish the project in limit budget on time.
The objective of the research is to estimate project completion probability and to establish the project expected profit model. It includes minimizing project direct cost, estimating project completion probability, establishing project expect profit model, and risk response plan. First, PSO heuristic algorithm is used to get the minimum project direct cost. Project completion probabilities are estimated by Monte Carlo Simulation. The time and duration based on PSO method is compared with the probability matrix derived from Monte Carlo Simulation. Usually the on-time project completion rate can be increased by adding more project budget. This is justified based on the expected profit equation from the research. It is shown from the case study that the proposed research model and profit equation can assist the project manager in making better decision on project crashing.

並列關鍵字

Time-cost trade-off problem ； Probability matrix ； Expect profit model

參考文獻

3.吳育周，運用基因演算法規劃資源受限專案排程問題之探討，崑山科技大學企業管理研究所碩士論文，民國94年。

1.Abbasi, G. Y. and Mukattash, A. M. (2001). “Crashing PERT networks using mathematical programming” International Journal of Project Management 19(3), 181-188

2.Abdalla, M. Odeh and Hussien, T. Battaineh (2002). “Causes of delay in large construction projects” International Journal of Project Management 20, 67-73

3.AbouRizk, S. M. and Wales, R. J. (1997). “Combined discrete-event/continuous simulation for project scheduling” Journal of Construction Engineering and Management 123(1), pp.11-20

6.Chatterjee A., Siarry P. (2006). “Nonlinear inertia weight variation for dynamic adaptation in particle swarm optimization” Computers & Operations Research 33, 859-871