Abstract
One of the assumptions in PERT is the possibility to apply central limit theorem (CLT) to approximate path duration times with standard normal distribution. However, CLT presumes certain conditions in order to be correctly applied. This issue is often not adressed in related literature. The aim of this paper is to examine the conditions under which PERT can accordingly be applied. We consider the conditions and aspects of their practical application in order to verify the admissibility of CLT for given activity time distributions. Lindeberg-Feller condition turned out to be the simplest technique to verify that results of a PERT analysis are free from problematic CLT application. We also summarize other issues with current probabilistic project evaluation and propose a chance constraint optimization model for probabilistic project analysis.
| Original language | English |
|---|---|
| Title of host publication | Project Management Development - Practice and Perspectives Conference Proceedings |
| Pages | 114-127 |
| Number of pages | 14 |
| Publication status | Published - 2019 |
Fields of science
- 303026 Public health
- 305909 Stress research
- 102 Computer Sciences
- 102006 Computer supported cooperative work (CSCW)
- 102015 Information systems
- 102016 IT security
- 502007 E-commerce
- 502014 Innovation research
- 502030 Project management
- 501016 Educational psychology
- 602036 Neurolinguistics
- 501030 Cognitive science
- 502032 Quality management
- 502043 Business consultancy
- 502044 Business management
- 502050 Business informatics
- 503008 E-learning
- 509004 Evaluation research
- 301407 Neurophysiology
- 301401 Brain research
JKU Focus areas
- Digital Transformation