Page 61 - Contributed Paper Session (CPS) - Volume 2

CPS1428 Enobong F. U.

Burr XII distribution and its application in PERT
Enobong Francis Udoumoh
University of Agriculture, Makurdi, Nigeria

Abstract
Project Evaluation and Review Technique (PERT) in conjunction with Monte
Carlo simulation is no doubt a useful and an acceptable tool for project
network analysis. However, a common challenge in practice is the inability to
confidently select appropriate input distribution to represent activity duration
owing to scarcity of project data. In this research, Burr XII distribution is
introduced as an input duration distribution in the analysis of project network.
Some desirable properties that qualify Burr XII distribution as input duration
distribution are presented. The classical quantile estimation approach is
adopted to estimate activity parameters based on judgmental estimates. To
illustrate the procedure, activity duration data from a survey of bore hole
drilling projects in Benue State, Nigeria was used. Initial results based on
goodness of fit test shows that Burr distribution is well suited to model activity
duration. A program written in MATLAB codes is used to Monte Carlo the
project network to estimate the Project Completion Time (PCT). Further results
revealed that the new approach which uses Burr XII distribution performs well
especially when the activity distribution is heavily tailed to the right.

Keywords
Activity Duration; Quantile Estimation; Project

1. Introduction
Management of projects has become increasingly complex owing to the
sophisticated nature of our environment and management systems. This has
led to the adoption and formulation of efficient management techniques for
effective management of projects. The Critical Path Method (CPM)/Project
Evaluation and Review Technique (PERT) is one of the management techniques
which has been use extensively in project management,Tavares (2002),
PMBOK (2008). The originators of PERT (Malcom et al. 1959) assumed that
project activity time follows the generalized beta distribution with probability
density function
Γ(+) (−) −1 (−) −1
(0 = ; < < , , > 0 (1)
Γ()Γ() (−) +−1

50 | I S I W S C 2 0 1 9