A model of correlated team behaviour in a software development environment

In today's highly competitive software development environments, accurately estimating software duration and cost can often mean the difference between project success or failure. Traditional software development estimation techniques often assume that software development teams operate independently from task to task. This assumption allows task covariances to be ignored. However, there is ample evidence that, in practice, the behaviour of a software development team over a project life-cycle has strong "memory" of previous tasks and external influences. This means that the accuracy of software development estimates may suffer due to this simplifying assumption of no-correlation. We present a model that we use to describe correlated team behaviour in a software development environment.