**1. Introduction**

Project Management is the discipline to manage, monitor and control the uncertainty inherent to projects. Project managemen<sup>t</sup> processes are used to monitor and control the progress of projects in order to reduce the uncertainty, and each such process requires effort from the project manager and her team. The academic literature has been overwhelmed by research studies in project managemen<sup>t</sup> and control, and many of them focus on the construction of the project baseline schedule to assess the project risk and to monitor the performance of a project in progress. The combination of these three dimensions—schedule, risk and control—is often referred to in the literature as *dynamic scheduling* [1,2] or *integrated project management and control* [3].

This paper starts with the observation that the relation between managerial effort and the ability to reduce the project uncertainty lies at the heart of many research studies, although this relation is

often not explicitly mentioned. Especially in some research papers that rely on the concept of *entropy* as a way to express that projects have the natural tendency to move to a state of disorder, authors have referred to the relation between entropy (uncertainty) and energy (effort). They have proposed different entropy measures to enable the project manager to better predict the project uncertainty and eventually reduce it by taking better actions. This concept of entropy is—to the best of the authors' knowledge—not widely used in the previously mentioned dynamic scheduling studies; however, it is believed that it sheds an interesting light on the project managemen<sup>t</sup> domain and opens ways to look at the dynamic scheduling literature (schedule/risk/control) in a fundamentally different way.

The current study reviews the research on entropy in project managemen<sup>t</sup> and proposes a new way to accurately estimate project uncertainty to improve project forecasting and decision-making. This paper first elaborates on the link between the traditional dynamic scheduling literature and the much less investigated concept of entropy in project managemen<sup>t</sup> and argues that entropy is an ideal concept to measure project uncertainty. Then, it will be shown that, in order to reduce a project's entropy, forecasting and estimates are crucial for a project manager and her team to make well-informed decisions. Then, finally, a new so-called *calibration method* is proposed that should help project managers to better quantify the project uncertainty by providing better estimates for the activity durations. Such calibration procedures are relatively new in the literature, since they rely on a combination of statistical data analysis and the correction for human biases.

The paper is organized in the following sections. Section 2 reviews the most important studies on entropy for managing projects that have been used as an inspiration for the current research study. Based on this, the section also explains the basic idea of calibrating project data to better estimate project uncertainty, which constitutes the main theme of our study. In Section 3, two currently known data calibration methods from literature are then briefly reviewed, as they will be used as foundations for a new third calibration method taking the shortcomings of the existing methods into account. This new so-called *statistical partitioning heuristic* is discussed in Section 4. Section 5 presents the results of a computational experiment on a set of 83 empirical projects (mainly construction projects) from a known database. The section shows that the statistical partitioning heuristic outperforms the two other procedures, but also discusses some limitations that can be used as guidelines for future research. Finally, Section 6 draws conclusions and highlights some potentially promising future research avenues.
