An Information-Theoretic Model of Project Schedule Overruns Caused by Task Rework: A Case for Newspeak

Many large-scale, complex engineering systems experience significant cost and schedule overruns during their developments. There are many factors that contribute to these overruns, including increased system complexity, task rework, and the in-ability to exhaustively test all states and configurations of a given system, which often leads to re-design efforts. In this work, we focus specifically on task rework and its impact on project schedule overruns. We demonstrate that heavier tail phenomena present in large-scale program development duration can be caused by task rework. Within this context, we hypothesize that one cause of task rework in a project development effort is misinterpretation of task instructions. We develop a computational framework for estimating the information content of a set of task instructions and the expected time to task completion. This reveals that heavier tailed duration phenomena present in many large-scale project development efforts can arise due to task rework caused by misinterpretation.