Fitting activity distributions using human partitioning and statistical calibration
Name:
Publisher version
View Source
Access full-text PDFOpen Access
View Source
Check access options
Check access options
Publication type
Journal article with impact factorPublication Year
2019Journal
Computers and Industrial EngineeringPublication Volume
129Publication Issue
MarchPublication Begin page
126Publication End page
135
Metadata
Show full item recordAbstract
Many project management and scheduling studies have modelled activity durations as a range of values to express the stochastic nature of projects in progress. A wide variety of simulation models have been proposed that all rely on pre-defined statistical probability distributions for the durations of project activities. Ideally, these distributions reflect the real stochastic nature of the activities to assure that the simulations imitate the expected reality in the best possible way. However, the distributions are often selected ad hoc, relying on a class of distributions that are often used in the statistical literature, but without having much links with the features of real projects. Recently, a calibration method has been proposed in literature and validated on a set of 24 projects that makes use of real project data to derive realistic statistical distributions. This paper builds further on the validation of this calibration method in three different ways. First, the procedure is now successfully used on a set of 125 projects (for which 83 could be used for the final analysis) from different sectors. Secondly, the procedure has been extended with a partitioning step performed by humans with experience in the particular project. Finally, some procedural extensions have been proposed to test the necessity of each step of the procedure.Keyword
Project Management, Empirical Data, Activity Durations, Distribution Fitting, Parkinson Distribution, Lognormal Distribution, Project Partitioning, Managerial Criteria, Risk ProfilesKnowledge Domain/Industry
Operations & Supply Chain Managementae974a485f413a2113503eed53cd6c53
10.1016/j.cie.2019.01.037