A Framework for Assessing Safety Argumentation Confidence

Software applications dependability is frequently assessed through degrees of constraints imposed on development activities. The statement of achieving these constraints are documented in safety arguments, often known as safety cases. However, such approach raises several questions. How ensuring that these objectives are actually effective and meet dependability expectations? How these objectives can be adapted or extended to a given development context preserving the expected safety level? In this paper, we investigate these issues and propose a quantitative approach to assess the confidence in assurance case. The features of this work are: 1) fully consistent with the Dempster Shafer theory; 2) considering different types of arguments when aggregating confidence; 3) a complete set of parameters with intuitive interpretations. This paper highlights the contribution of this approach by an experiment application on an extract of the avionics DO-178C standard.