A Model-Driven Approach for Evaluating Traceability Information

A traceability information model (TIM), in terms of requirement traceability, describes the relation of all artifacts that specify, implement, test, or document a software system. Creating and maintaining these models takes a lot of effort, but the inherent information on project progress and quality is seldom utilized. This paper introduces a domain-specific language (DSL) based approach to leverage this information by specifying and evaluating company-or project-specific analyses. The capabilities of the Traceability Analysis Language (TAL) are shown by defining coverage, impact and consistency analysis for a model according to the Automotive Software Process Improvement and Capability Determination (A-SPICE) standard. Every analysis is defined as a rule expression that compares a customizable metric's value (aggregated from the TIM) against an individual threshold. The focus of the Traceability Analysis Language is to make the definition and execution of information aggregation and evaluation from a TIM configurable and thereby allow users to define their own analyses based on their regulatory, project-specific, or individual needs. The paper elaborates analysis use cases within the automotive industry and reports on first experiences from using it.