Entropy and Risk Management
By Mithun Sridharan
In this article, I shall be dealing with the topic – Project Risk Management and demonstrate how one could find mathematical equivalents for an abstract methodology. Any project, large or small is associated with expected and unexpected problems. The analogy mentioned above could be derived from the Second Law of Thermodynamics. The Second law of thermodynamics deals with a concept : Entropy. Entropy, in short, is the amount of disorderliness of the system. Entropy is also a measure on the information contained in an system. In information technology, entropy is considered as the amount of uncertainty in an given system. This has a defined relation, “As the amount of information increases, the disorderliness of a system (entropy) decreases”.
Considering our project management scenario as a system, one could come up with a model system, comprising of micro and macro phases, each having a certain amount of entropy. Considering the information to be inversely proportional to the entropy (okay, almost inversely!), means to increase the information, through careful planning communication and monitoring the project progress would increase the information available to a project manager. This way, the entropy decreases. As uncertainty is proportional to risk, decreasing the entropy or increasing the information available, one could decrease surprises.
Projects typically follow a certain distribution, which is unique to the project as the project itself. This lends well to modeling the project using one of the several proven mathematical models of Information theory and study the progress to follow the distribution. If the project exhibits a pattern, then using the statistics, one could directly plug the values such as cost/schedule variances and/or other indites to monitor, track and optimistically, predict the project behavior.
The situation is much more complicated than the classical project management from the technological advancements, which are both a boon and a bane. With the Internet and Networking technology, the amount of information available has exponentially increased over the years, thereby increasing the entropy of the systems. Selective information processing and the indices available from the existing project management literature is no longer adequate. I’ll provide my standard IT example here. IT projects are technically much more demanding than the projects from the classical industry standard projects. We are talking about professional project management efforts in a very highly complicated and abstract technology, which has several abstract interfaces that could take any form depending upon the sub set of parameters that influence the outcome at any particular moment. Thus, studying entropy of such systems and observing the behavior to fit any standard distribution would help you close the gap between the existing high level of abstraction and the level of information to properly direct the project effort.
Compliance is another project in IT projects, if not for no other reason, than the trotting status of IT projects. Technology has traditionally kept building itself over the existing infrastructure, but IT projects have proven to refute this standard behavior. Thus, compliance to the proven methodologies may need necessarily not prove a good project standing. An IT project may fall short of the expectations a day before going live. The kind of data available and its consumption as well as decision making also has a definitive complexity, which multiplies the project risk. Add to it, the fact that the computer programs have a life cycle of their own and that these programs “evolve” with time, its almost next to impossible to assign one particular behavior to IT projects. Additionally, the project team working on the program development may or may not outlive the software life time. There is a “human factor” to these projects, which is not time proven. We do not have the “legacy” as with other application areas.
Segmenting the project into micro and macro states, studying the individual parts and gathering the parameters of these and analyzing their fit in the “whole” such as a fine grained approach, without losing the information from the “coarse grained” paradigm would complement each other in projects of complexity as those mentioned above.
Mithun Sridharan works for the Independent Software Vendor (ISV) Engineering team for Sun Microsystems and is based in Walldorf, Germany. His interests include project management, poetry, general management, philosophy and chess. A PMP and an ardent blogger, he writes extensively on project management in his blog: Mithun’s Memoirs.