ALBERT

Description: Most construction managers use deterministic scheduling techniques to plan construction projects and estimate their duration. However, deterministic techniques are known to underestimate the project duration. Alternative methods, such as Stochastic Network Analysis, have rarely been adopted in practical contexts as they are commonly computer-intensive, require extensive historical information, have limited contextual/local validity and/or require skills most practitioners have not been trained for. In this paper, we propose some mathematical expressions to approximate the average and the standard deviation of a project duration from basic deterministic schedule information. The expressions’ performance is successfully tested in a 4100-network dataset with varied activity durations and activity durations variability. Calculations are quite straightforward and can be implemented manually. Furthermore, unlike the Project Evaluation and Review Technique (PERT), they allow drawing inferences about the probability of project duration in the presence of several critical and subcritical paths with minimal additional calculation.

Description: Just like any physical system, projects have entropy that must be managed by spending energy. The entropy is the project’s tendency to move to a state of disorder (schedule delays, cost overruns), and the energy process is an inherent part of any project management methodology. In order to manage the inherent uncertainty of these projects, accurate estimates (for durations, costs, resources, …) are crucial to make informed decisions. Without these estimates, managers have to fall back to their own intuition and experience, which are undoubtedly crucial for making decisions, but are are often subject to biases and hard to quantify. This paper builds further on two published calibration methods that aim to extract data from real projects and calibrate them to better estimate the parameters for the probability distributions of activity durations. Both methods rely on the lognormal distribution model to estimate uncertainty in activity durations and perform a sequence of statistical hypothesis tests that take the possible presence of two human biases into account. Based on these two existing methods, a new so-called statistical partitioning heuristic is presented that integrates the best elements of the two methods to further improve the accuracy of estimating the distribution of activity duration uncertainty. A computational experiment has been carried out on an empirical database of 83 empirical projects. The experiment shows that the new statistical partitioning method performs at least as good as, and often better than, the two existing calibration methods. The improvement will allow a better quantification of the activity duration uncertainty, which will eventually lead to a better prediction of the project schedule and more realistic expectations about the project outcomes. Consequently, the project manager will be able to better cope with the inherent uncertainty (entropy) of projects with a minimum managerial effort (energy).