Modelling time-constrained software development

Commercial pressures on time-to-market often require the development of software in situations where deadlines are very tight and non-negotiable. This type of development can be termed ‘time-constrained software development.’ The need to compress development timescales influences both the software process and the way it is managed. Conventional approaches to modelling tend to treat the development process as being linear, sequential and static. Whereas, the processes used to achieve timescale compression in industry are iterative, concurrent and dynamic. That is, they replace the notion of ‘right-first-time’ with one of ‘right-on-time.’ In this paper we propose a new modelling technique, called Capacity-Based Scheduling (CBS), to control risk across a portfolio of time-constrained projects. We show how schedule constraints can be modelled in order to predict the consequences of alternative plans and control schedule risk across a portfolio of time-constrained projects.