An Analytical Model of Requirements Selection Quality in Software Product Development

In market-driven requirements engineering a crucial decision is whether or not to select a requirement for implementation in the next software release. The selection process is part of the release planning process and can be viewed as a series of refinement steps resulting in either rejection or propagation to the next step. This paper presents an analytical model of the selection process which takes the quality of the decision-making into account. The model is a network of queues with multiclass jobs corresponding to requirements of different quality. The analytical model can act as a baseline for simulation of more realistic models where no analytical solution is possible. Further work includes empirical assessment of the model and further experiments using simulation. The long-term objective of the presented research is to provide means for practitioners to optimize the selection process in terms of minimizing wasted effort.