QRA: A Quality Requirements Analysis Approach for Service Systems

Although quality requirements (QRs) are considered as being of crucial importance in today's service-oriented systems, existing approaches almost exclusively deal with QRs from the service providers' perspective (namely late QRs). The motivation for this paper is to address the analysis of QRs from the service customers' perspective (namely early QRs). The work presented in this paper focuses on the way that early QRs may be modeled and evaluated and demonstrates this approach on a stock trading service system - a real-world practice taken from an international firm in the financial sector. We focus on three QRs that are critical to service systems and especially that of stock trading, namely performance, availability and security. We introduce a modeling paradigm that extends the well-known UML Activity Diagram by explicitly representing these three QRs as an integral part of business process modeling and augments this modeling with quantitative and qualitative reasoning that together provide the means for enhanced decision making by service customers.