Extended Knowledge-Centric Traceability ManagementMethod improvement for the management of relationships between system artefactshttps://repo.valu3s.eu/method/extended-knowledge-centric-traceability-managementhttps://repo.valu3s.eu/@@site-logo/logo_valu3s_green_transparent.png
Method improvement for the management of relationships between system artefacts
To manage the relationships between system artefact by taking advantage of ontologies and semantic information, further supporting advanced traceability project configuration and automating trace discovery and verification.
Knowledge-centric traceability management is a method that exploits knowledge bases (ontologies) for traceability purposes, e.g., to represent the system artefacts to trace and relationships between them, or to discover or verify relationships. This method has been improved to support further activities of the traceability management process and the integration with other engineering activities from which system artefact information can be collected for traceability purposes.
Enhanced traceability management process. The process for using Knowledge-centric traceability management has been refined and extended to better support traceability management activities. The main improvements can be summarised as follows:
Definition of maps for the traceability projects, to show an overview of the configuration, modules, and data sources of a project. Maps have also been associated to the individual modules of a traceability project.
Semantic configuration, to specify aspects such as trace types and trace information retrieval configuration.
Advanced trace information retrieval configuration, currently used to suggest traces and identify suspect ones. The possible configurations affect similarity threshold, NLP configuration for retrieval, and the method (inclusion or similarity).
As follow-up improvements, traces can be discovered (suggested ones) and evaluated (suspect ones). A semantic distance score is provided. As a result, traces can be classified as Suspect, Suggested, and Consistent.
Configuration of trace discovery and evaluation according to semantic information (specification patterns and semantic clusters).
Trace history management, to keep track of trace lifecycle.
Graph generation for impact analysis, with different options for graph configuration.
Further mapping of the underlying data model to the models of other tools, e.g., DOORS for traceability information exchange.
Determination of new report generation possibilities, according to the information managed about a traceability project.
Some improvements have already been referred to in a publication [EKCT1].
Advanced automated traceability management
Method that can be applied for any system artefact type
The method could be improved by integrating it with other approaches, e.g., model-based traceability management
No empirical evidence of cost-effectiveness
[EKCT1] de la Vara, J.L., García, A., Valero, J., Ayora, C.: Model-Based Assurance Evidence Management for Safety-Critical Systems. Software and Systems Modeling (accepted paper), 2022