A Model for Quality Optimization in Software Design Processes

The main objective of software engineers is to design and implement systems that implement all functional and non-functional requirements. Unfortunately, it is very difficult or even generally impossible to deliver a software system that satisfies all the requirements. Even more seriously, failures in fulfilling requirements are generally detected after the realization of software systems. This is because design decisions are mostly taken based on estimations, which can turn out to be wrong at a later stage in the design process. Switching to different design alternatives at a later stage can be very difficult since this may demand drastic changes in design and also may increase project time and costs. In this paper a model is proposed for modeling and tracing design processes with respect to the selected design alternatives. Based on the model, two algorithmic definitions of design strategies are given, which enable software engineers to optimize design decisions with respect to quality and resource constraints.