Semi-quantitative Simulation Modeling of Software Engineering Process

Abstract

Software process simulation models hold out the promise of improving project planning and control. However, purely quantitative models require a very detailed understanding of the software process, i.e. process knowledge represented quantitatively. When such data is lacking, quantitative models impose severe constraints, restricting the model’s value. In contrast, qualitative models display all possible behaviors but only in qualitative terms. This paper illustrates the value and flexibility of semi-quantitative modeling by developing a model of the software staffing process and comparing it with other quantitative staffing models. We show that the semi-quantitative model provides more insights into the staffing process and more confidence in the outcomes than the quantitative models by achieving a tradeoff between quantitative and qualitative simulation. In particular, the semi-quantitative simulation produces a set of possible outcomes with the ranges of real numeric values. The semi-quantitative model allows us to determine the solution boundaries for specific scenarios under the conditions of limited knowledge.