|Authors||M. Sabetzadeh, D. Falessi, L. Briand and S. Di Alesio|
|Title||A Goal-Based Approach for Qualification of New Technologies: Foundations, Tool Support, and Industrial Validation|
|Afilliation||Software Engineering, Software Engineering|
|Publication Type||Journal Article|
|Year of Publication||2013|
|Journal||Reliability Engineering and System Safety (RESS)|
New technologies typically involve innovative aspects that are not addressed by the existing normative standards and hence are not assessable through common certification procedures. To ensure that new technologies can be implemented in a safe and reliable manner, a specific kind of assessment is performed, which in many industries, e.g., the energy sector, is known as Technology Qualification (TQ). TQ aims at demonstrating with an acceptable level of confidence that a new technology will function within specified limits. Expert opinion plays an important role in TQ, both to identify the safety and reliability evidence that needs to be developed, and to interpret the evidence provided. Since there are often multiple experts involved in TQ, it is crucial to apply a structured process for eliciting expert opinions, and to use this information systematically when analyzing the satisfaction of a technology's safety and reliability objectives. In this article, we present a goal-based approach for TQ. Our approach enables analysts to quantitatively reason about the satisfaction of a technology's overall goals and further to identify the aspects that must be improved to increase goal satisfaction. The approach is founded on three main components: goal models, expert elicitation, and probabilistic simulation. We describe a tool, named Modus, that we have developed in support of our approach. We provide an extensive empirical validation of our approach through two industrial case studies and a survey.