Impact of limited data availability on the accuracy of project duration estimation in project networks

Published: 25 July 2022| Version 2 | DOI: 10.17632/bjfdw6xbxw.2
Contributors:
Naimeh Sadeghi,

Description

This database includes simulated data showing the accuracy of estimated probability distributions of project durations when limited data are available for the project activities. The base project networks are taken from PSPLIB. Then, various stochastic project networks are synthesized by changing the variability and skewness of project activity durations. Number of variables: 14 Number of cases/rows: 330028 Variable List: Network ID: ID of the synthesized network #Activities: Number of activities in the network, including start and finish activities Net var: Variance of the activities in the network (“Net var” can be either high, low, medium or rand, where rand shows a random combination of low, high and medium variance in the network activities.) Net skew: Skewness of the activities in the network (“Net skew” can be either right, left, None or rand, where rand shows a random combination of right, left, and None skewness in the network activities) Number of arcs: Number of arcs in the activity on node project network Benchmark mean: Mean project duration in the benchmark stochastic project network sample size: Number of sampled data used for the experiment resembling limited data condition Fitted distribution type: Distribution type used to fit on sampled data Average fitting KS: Kolmogorov–Smirnov(K–S) test value averaged among all K-S values resulted from the fitting process using the maximum likelihood method for network activities #Criticals: Number of critical activities in the project network, averaged among Monte Carlo simulation runs Mean: Mean of project duration distribution for the experiment Variance: Variance of project duration distribution for the experiment KS: Kolmogorov–Smirnov test comparing benchmark distribution and project duration distribution of the experiment MAPE: Mean absolute percentage error comparing benchmark distribution and project duration distribution of the experiment

Files

Steps to reproduce

Data can be reproduced using the python code available in the folder with the same name.

Institutions

K N Toosi University of Technology Faculty of Civil Engineering

Categories

Uncertainty, Project Management, Project Scheduling, Probability Distribution

License