Every process model used by software industry has different phases including requirement engineering. This is the crucial phase as it is preceded by other phases and provides valuable inputs to the design phase. Risk assessment made in this phase can help avoid wastage of time, effort, cost and budget overruns and even missed delivery deadlines. Traditionally risks are analyzed in terms of technical aspects like failures in the working system, unavailability of certain services, and fault intolerances to mention few. The identified risks are used to have countermeasures. However, it causes the life cycle of the system to be repeated right from the requirements engineering. On the contrary, risk analysis in the requirements engineering phase can prove fact that a stitch in time saves nine. Therefore early detection of risks in the system can help improve efficiency of software development process. Goal-oriented risk assessment has thus gained popularity as it is done in the requirements analysis phase. Stakeholder interests are considered to analyze risks and provide countermeasures to leverage quality of the system being developed. In this paper, a formal framework pertaining to Tropos goal modelling is enhanced with quantitative reasoning technique coupled with qualitative ones. Towards this end we used a conceptual framework with three layer such as asset layer, event layer and treatment layer. We used a case study project named Loan Origination Process (LOP) to evaluate the proposed framework. Our framework supports probability of satisfaction (SAT) and denial (DEN) values in addition to supporting qualitative values. The Goal-Reasoning tool is extended to have the proposed quantitative solution for risk analysis in requirements engineering. The tool performs risk analysis and produces different alternative solutions with weights that enable software engineers or domain experts to choose best solution in terms of cost and risk. The results revealed the performance improvement and utility when compared with an existing goal-driven risk assessment approach.
Risk Analysis for Collaborative Systems during Requirements Engineering (S)
