New publication: Statistical Project Control

New publication on Statistical Project Control

Finally, after years of hard work, our first paper on Statistical Project Control has been published in the flagship journal Omega - The International Journal of Management Science.

The peer-reviewed process

An important aspect of our research is to provide valid answers on the criticism we receive in the peer review process of our paper submissions. The importance of the peer review process cannot be underestimated. It is an essential and critical part of the functioning of the scientific community, of quality control, and the self corrective nature of science. To that respect, the non-peer reviewed articles are nothing more than a way to spread some ideas to a bigger audience, which is an essential goal of research. But only a peer review mechanism has the necessary component of the essential quality control of research.

The current paper on Statistical Process Control is an excellent illustration of how hard this process can be, but also how it finally results in a much more improved version of the initial manuscript. Our paper has been initially submitted on 10/01/2012, and three additional revisions were necessary, leading to literally almost 100 extra pages of material and terrabytes of additional data to run new tests, before it could be accepted. Finally, the paper has been accepted in the third revision round in June 2014 under the title "Setting tolerance limits for statistical project control using earned value management" authored by Jeroen Colin and Mario Vanhoucke.

The peer-reviewed paper process: click on the picture to access the result

This process is often unknown or not well understood by practitioners (non-academics) who write articles in the more business-oriented journals without much rework. Therefore, you must know that every academic paper is the result of years of hard work, literally months of testing on fast computers using a sound and proven methodology, additional months to years of working on the revisions and of course also a little bit of luck. Every little detail matters and the smallest ambiguity can lead to a rejection. There's no need to mention that we are proud on the outcome.

The paper

In the latest version of the book "The Art of Project Management: A Story about Work and Passion", a short overview of the research paper has been written as follows: Project control systems must indicate the direction of change in preliminary planning variables compared with actual performance. In case the project performance of projects in progress deviates from the expected planned performance, a warning must be indicated by the control system as a trigger to take corrective actions.

Research

A new Statistical Project Control (SPC) approach based on the principle of statistical process control charts is presented in order to improve the discriminative power between normal and abnormal project progress situations. Based on the existing and commonly known Earned Value Management (EVM) metrics, the project control charts will have an improved ability to trigger actions when variation in a project's progress exceeds certain predefined thresholds.

Methodology

A large number of simulation experiments has been set up using P2 Engine running on Ghent University’s super computer infrastructure, leading to gigabytes of data in order to test the ability of the statistical project control charts to discriminate between random and assignable variation. An intensive analysis of the generated data is done to compare the use of statistical project control limits with traditional earned value management thresholds and to validate their power to report warning signals when projects run into danger.

Results

The results of the computational experiments show that:

  • The use of SPC outperforms the best practices in EVM.
  • The Earned Schedule (ES) approach performs better than the traditional EVM approach.
  • A combined use of X-charts and XR-charts allows to detect a variety of project problems.
  • An extended multi-variate analysis control approach leads to control efficiency improvements.