專案完成時間延遲原因之研究

On the Research of the Delay Causes of the Network Completion Time

褚文明(Wen-Ming Chu)

《興國學報》 9期 (2008/01) Pp. 45-57

https://doi.org/10.29950/JHKUM.200801.0004

摘要

根據Coppendale（1995：35-38）及Chatzoglou and Macaulay（1996：173-183）的研究，平均僅約10%－15%的大型專案能夠如期完成，餘多數均會發生執行延遲的現象，這不僅造成專案完成時間上的拖延，更會讓企業經營成本的負荷增加。在實務上，專案所有的作業時間均為隨機變數，因此確定性專案網路的分析工具（例如CPM）已無法適用，而計畫評核術（PERT）對大型專案網路的時間估算，亦會出現相當大的誤差，因此專案完成時間無法獲準確的估測，若無適當的趕工，往往會有延遲的情況發生。本文即針對專案延遲的現象做研究，針對專案執行的過程做系統的分析，探討出專案延遲的原因，並發現主要是源於「最長路徑偏差（LPB）」及「專案網路路徑相依」兩現象所造成，論文中並針對該兩現象以實例來作說明。

關鍵字

計畫評核術 ； 隨機性專案網路 ； 專案網路

並列摘要

There is evidences from Coppendale (1995: 35-38)及 Chatzoglou and Macaulay (1996: 173-183), which indicate that for large projects, only approximately 10~15% of them finish their works on time. The majority of them overrun not only their planned schedule but also their project budget. Since all the activities of the networks are practically random variable, the analysis tools used in the deterministic network, do not work any more in the stochastic network. The most popular Project Evaluation and Review Technique (PERT) was applied for the approximation of the network completion time, but it would cause considerably inaccuracy in the approximation and it is usually need work overtime for project to avoid the delay. This paper tries to deep into the caused reasons of the project delay through the system analysis on the project running executions, and found out two of the principal causes: ”The Longest Path Bias (LPB)” and ”The dependence between paths in the network”. We will use real instance example to explain how their effects in the networks.

並列關鍵字

PERT ； Stochastic network ； Project network

參考文獻

Adlakha, V. G.,Kulkarni, V. G.(1989).A classsified bibliography of research on stochastic PERT networks: 1966-1987.Information Systems and Operational Research.3(27),272-296.

Bigelow, C. G.(1962).Bibliograghy on project planning and control by network analysis: 1959-1961.Operations Research.10,728-731.

Burt, J. M.,Garman, M. B.(1971).Conditional Monte Carlo: A simulation technique for stochastic network analysis.Management Science.3(18)

Chatzoglou, P.,Macaulay, L.(1996).A review of existing nodels for project planning and estimation and the need for a new approach.International Journal of Project Management.14,173-183.

Coppendate, J.(1995).Manage risk in product and process development and avoid unpleasant surprises.Engineering Management Journal.February,35-38.