Provenance and Traceability. The ability to relate the many artefacts stored within the different tools is an essential feature of the INTO-CPS tool chain. This feature has three different abilities: the ability to logically relate artefacts contained in the many tools, such as relating a requirement to a model that implements it, this we refer to as traceability; the ability to record the temporal links between entities produced dur- ing the development, such as connecting simulation results and the tools and models that created them, this we term the provenance; and finally the ability to query this stored data to produce useful informa- tion for the various stakeholders of the project. The success of the provenance and traceability activities will depend on the tool support, which begins in year 2 of the project, but more so on a solid on- tology of the INTO-CPS artefacts and relations that will be tracked. The ini- tial work describing these elements is included in Appendix A of [FGPP15a] and the interested reader is directed there to find more details. Initial explo- ration of how this can be incorporated in the tool suite has been carried out [Han16].
Provenance and Traceability methods for machine-assisted recording and mainte- xxxxx of links between models, multi-models and other design artefacts. The guidelines are underpinned by a common concept base, and are supported by a growing set of pilot studies that can serve as benchmarks for the methods and tools and as illustrations for the tool chain’s capabilities. The first release of the new multi-modelling tool chain was scheduled for late in Year 1. Our priority in WP3 was therefore to lay foundations for guidelines that will emerge from experience with the tool chain, and to provide specifications for functionality required of tools in the areas of Design Space Exploration (DSE) and traceability. Our specific objectives were therefore as follows: To survey the state of the art in multi-modelling methods, so that the project could identify promising techniques and tools and position its own contributions. This survey is reported in Section 3. To review the workflows in the case study partners and identify a candidate initial workflow against which the INTO-CPS tool chain could be validated once released. This was undertaken in Task 3.1 and is reported in Section 2.1. To specify the Design Space Exploration (DSE) support that would ultimately be required in the toolchain. This was undertaken in Task 3.2 and is reported in Section 2.2. The specification will be implemented in Task 5.1, for which Year 1 progress is reported in Deliverable 5.1a [GHJ+15]. To specify the traceability and provenance functionality that the toolchain should support. This would be implemented in Task 4.4 from Year 2 onwards. This was undertaken in Task 3.3 and is reported in Section 2.3. Details of the support for provenance features are given as Appendix A. The specification will be implemented in Task 4.4 from Year 2 onwards. To prepare a common concept base for the project, and deliver a first set of guide- lines for the construction of SysML models for multi-modelling given different entry points, domain knowledge and previous multi-modelling experience. This was un- dertaken in Task 3.4 and is reported in Section 2.4. The concept base is reported in D3.1a Method Guidance 1 [FGPP15]. To document initial pilot studies that illustrate various properties of CPSs and features of the INTO-CPS technologies. This was undertaken in Task 3.5 and the initial pilot studies are described in reported in Section 2.5.
Provenance and Traceability. The first year of provenance and traceability study has concentrated on exploring the both the relations that will be important to the record within the INTO-CPS tool chain and also potential standard notations and specifications that may be used to represent them. Two distinct specifications have been identified to form the foundations for the the provenance and traceability concepts within INTO-CPS, these specifications have the benefits of both being open and free to use, and, in case of OSLC, have a wide base of industry support. The W3C PROV 2 model provides support for recording the temporal relations between activities, entities and agents within a process. This supports the recording of, for example, links between simulation results and the models, platforms and configurations that produced them, this is important when generating documentation as part of a certification effort. To compliment this the relations specified by the OSLC 3 provides support for recording logical relations between objects within a data set. So OSLC allows the linking of, for example, a submodel and a requirement that it is designed to satisfy, or from a simulation result to the requirement it provides evidence for. The key concepts of both PROV and OSLC are presented in Appendix A of this document along with the proposed provenance and traceability ontology, which will form the basis for data recorded by the various tools in the INTO-CPS tool chain. This appendix also contains example applications of both PROV and OSLC being used to represent many of the document relations expected when using the INTO-CPS tool chain and workflows. 2xxxx://xxx.x0.xxx/TR/prov-overview/ 3xxxx://xxxx-xxxxxxxx.xxx
