Evaluation of Project Duration Uncertainty using the Dependency Structure Matrix and Monte Carlo Simulations
The sequence of tasks is vital to the development of any project. The order of tasks is influenced by the information flow among them. The dependency structure matrix (DSM) can be used to model information flow. However, the information used by the DSM, including task duration, time required for communication, and task overlap, can have uncertain values. The aim of this paper is to analyze the effect of uncertainty on the scheduling of tasks using the DSM. Monte Carlo Simulation is applied to represent uncertainty in time estimates for projects using DSM. Conventional project duration, normal project duration with communication times, and normal project duration containing natural overlapping were considered for project time estimations. Several distribution functions were used to represent the uncertainty. It is concluded that project duration has normal distribution behavior that is independent of the type of input parameter distribution functions; that mean values of the input parameters give a good estimation of mean project duration; that interval and inner interval arithmetic give overestimation and underestimation of project duration times respectively.
| Main Authors: | , , |
|---|---|
| Format: | Digital revista |
| Language: | English |
| Published: |
Escuela de Construcción Civil, Pontificia Universidad Católica de Chile
2015
|
| Online Access: | http://www.scielo.cl/scielo.php?script=sci_arttext&pid=S0718-915X2015000200010 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The sequence of tasks is vital to the development of any project. The order of tasks is influenced by the information flow among them. The dependency structure matrix (DSM) can be used to model information flow. However, the information used by the DSM, including task duration, time required for communication, and task overlap, can have uncertain values. The aim of this paper is to analyze the effect of uncertainty on the scheduling of tasks using the DSM. Monte Carlo Simulation is applied to represent uncertainty in time estimates for projects using DSM. Conventional project duration, normal project duration with communication times, and normal project duration containing natural overlapping were considered for project time estimations. Several distribution functions were used to represent the uncertainty. It is concluded that project duration has normal distribution behavior that is independent of the type of input parameter distribution functions; that mean values of the input parameters give a good estimation of mean project duration; that interval and inner interval arithmetic give overestimation and underestimation of project duration times respectively. |
|---|