Data Model and domain-specific ontologies implementation

The main results of task T2.2 “Semantic model and domain-specific ontologies presentations” are reported in this document, which is the logical continuation of the deliverable “D1.4 Harmonized set of semantic standards, ontologies, nomenclatures outcomes”.

The aim of D2.2 is to represent the main results related to data modelling, data collection and exchange specifications by providing the project and specifically the developers with several resources, artifacts, guides, etc. available on a public web framework. A significant effort was dedicated to the definition of data structures, nomenclatures and taxonomies, targeted for the specific topics of TRICK domain and services. The taxonomy of manufacturing operations in TC sector, reported in ANNEX B, is the most relevant, as it is an update of an outcome of the European project SET “Save energy in textile SMEs”, and collects inputs in particular on the Recycling phases from several TRICK project partners.

The requirements for supporting data collection along a highly fragmented supply chain, as described in D1.4, leaded us to focus on inter-company data exchanges and on their implementation through business documents that are the digital equivalent of the traditional paper based exchanges adopted by the companies, especially the smaller ones. For this reason, we have a strong attention on data exchanges standards and the effort to build upon the existing ones.

Starting from the state-of-the art reported in D1.4, the eBIZ specification framework was chosen as a basis for building the TRICK data model, improving its “eBIZ Document Factory” (see section 2.3) through its customization for TRICK and the development of additional features. Moreover, it was decided not to formalize the ontology in the classical RDF/OWL format, but to support it through an accessible online dictionary and to provide XML Schemas, in order to facilitate the work of developers in obtaining the corresponding data structures directly from an XML structure.

The approach followed for the replication in the Food sector is described in Chapter 4. The partners involved in the pilot at the time of writing are still working on the finalization of the two use cases under analysis; in parallel, the first analysis of the information and documents that must be exchanged was completed.

The analysis carried out to build the data model presented in this document was conducted starting from the TC sector and related pilots, identifying the common and transversal elements (see section 3.2) and highlighting which set of data could be mapped for the Food sector.Once the defined data model will be tested for the TC pilots, it will also be mapped on the specific features of Food pilots.

The Chapter 5 reports the preliminary approach of the Holonix Ecosystem Data manager adapted to handle the expected data flows and to serve the project specific uses cases, together with technical challenges and limitations. This preliminary analysis and all the results of T2.2 are the input for T2.3, which will use these outcomes to implement the data persistency layer for the TRICK platform.

The most important results of the Task 2.2 can be summarized as the following:
• The definition of the traceability data collection (see section 2.3.1) and of the sustainability assessment data collection (see section 2.3.2), with their related eBusiness documents, which provide the mapping of the TRICK domain and of the services’ specific topics. In fact, in order to have a more detailed analysis on the specific needs of the TRICK business services, also aspects not initially foreseen by the objectives of this deliverable were considered. An example is the preliminary data mapping on the TRICK services, reported in ANNEX E, and the consequent definition of a pattern for the services’ data support (see section 3.2.2) and the usage of the eBIZ elements in the TRICK services (see section 3.2.3).
• The instruments provided by the eBIZ Document Factory that are foreseen both for purposes of data structure design and maintenance as well as for on-line consultation of the delivered specifications for the developers.
• The Holonix Ecosystem Data Manager (EDM) adapted and customized to handle the expected data flows in a narrow specific domain like in the TRICK case.

Thanks to the approach and the choices made to build the TRICK ontology, described in Chapter 1, a standardization path may be followed to release a new version of eBIZ specification. Very important, in this perspective, is the relationship with the current normative evolutions about the digital product passport (DPP) that are considered in their potential of synergies with the TRICK outcomes, both as contributors to their definition as a best practice as well as potential clients of the future infrastructure.