Report on policy support for e-Procurement Version 1.0 (The Landscaping Document)

The Working Group for building consensus on the eProcurement ontology is made up of the following actors

A Review Group should be invited to provide an independent external review on the first full draft of the ePO. This will be done as part of the public comment period. The members of the Review Group will come from the same groups and communities as the members of the Working Group. Ideally, a member of the Working Group should not also be a member of the Review Group.

The OWL language is built upon the RDF standard. It is an ontology modelling language for describing RDF data. It allows for the strict definition of concepts and the complex relationships between them . The eProcurement Ontology should be expressed in OWL since the Common Data Model of the TED Semantic Web Service - in which the ePO will be implemented - is expressed in OWL.

SPARQL is a semantic query language. It is used to retrieve and manipulate data stored in RDF format.

Protégé is an open source ontology editor developed and maintained by Stanford University.

The CEN Workshop on business interoperability interfaces for public procurement in Europe (CEN WS/BII), established in March 2007, had the objective of providing a basic framework for technical interoperability in pan-European electronic transactions. It delivered a complete set of profiles covering both the pre-award and the post-award phases of the public procurement. The existence of these profiles and their associated semantic data models has been key in allowing disparate solutions to interoperate.

These profiles describe aspects throughout the whole procurement process such as notifications, the call for tenders, awarding and contracting. These profiles were implemented in several projects in Europe: the European Commission used them to build e-Prior, their open source solution for electronic invoicing and ordering, now also covering the pre-award phases; the PEPPOL community has also used them to create their own BIS specifications, resulting in a national-wide deployment of electronic invoicing in countries such as Norway, Denmark and Sweden, and other public administrations in Europe are currently basing their IT infrastructure and electronic procurement policies on deploying these standards e.g.,the National Health Service of the United Kingdom.

These profiles were updated in 2015 and examples of some profiles are listed below in Table 6 Examples of CEN BII Profiles.

These semantic models and their mappings to XML document exchange syntaxes, such as UBL and UN/CEFACT, should now be converted into knowledge to enable them to go a step further, by promoting a whole set of new functionalities such as searching for opportunities by sellers, comparing offers by buyers, getting statistical data, or improving the control and transparency in the electronic procurement procedures in the European Union.

In 2015, CEN established a new technical committee (TC) whose purpose is to develope standards to support and facilitate the electronic exchange of information in public procurement [_8]: CEN/TC 440. The technical committee will develop semantic data models, based on CEN/BII. TC/440 will closely collaborate with CEN/TC 434, a technical committee for the development of standards supporting European Electronic Invoicing [_9]. The work of CEN/TC 440 and TC 434 is closely related to the development of the ePO. Therefore synergies between CEN TC/440, TC 434 and the ePO should be developed as far as possible.

The Open Contracting Data Standard (OCDS) was developed for the Open Contracting Partnership (OCP) by the World Wide Web Foundation . The OCDS enables disclosure of data and documents at all stages of the contracting process by defining a common data model. It was created to support organisations to increase contracting transparency, and allow deeper analysis of contracting data by a wide range of users [_10].

The Open Contracting Data Standard (OCDS) is maintained using JSON Schema. Table 7 displays the main sections and common objects used in the schema.

The Open Contracting Data Standard cannot be directly reused in the ePO, because it is not an RDF vocabulary. It can however be used as an insight into all things that need considering during the modelling process as it is neatly structured and quite extensive. How it has developed its buyer URI could be analysed more in-depth.

Universal Business Language (UBL) has been designated by the European Commission as one of the first consortium standards officially eligible for referencing in tenders from Public Administrations and is freely available to everyone without legal encumbrance or licensing fees.

UBL is the result of an international effort to define a royalty-free library of standard electronic XML business documents, such as purchase orders and invoices.It is designed to plug into existing legal, business, auditing, and records management practices, eliminating the re-keying of data in existing fax and paper-based supply chains and being an entry point into e-commerce for SMEs [_12]. It is also used by nations around the world for implementing cross-border transactions related to sourcing (e.g. tendering), procurement (e.g. electronic invoicing), replenishment (e.g. managed inventory) and transportation (e.g. waybills and status).

The standard is the foundation for several European Public Procurement frameworks, including EHF (Norway) , Svefaktura (Sweden) , OIOUBL (Denmark) , e-Prior (European Commission DIGIT) , and PEPPOL [_13].

Universal Business Language provides a list of business objects expressed as reusable data components (e.g. address and payment) and common business documents (e.g. order and invoice), schemas for reusable data components and schemas for reusable business documents. Business objects from UBL that relate to the procurement field, include Invitation for Tender, Submission of Qualification Information and Awarding of Tenders. UBL Document Schemas related to e-Procurement include, for example, Call for Tenders. An example of these objects and how the relate, is described below.

Example

Business Object: Invitation to Tender

In this Business Object, i.e. the Invitation to Tender process, the Document Schema Call for Tenders is used. The Call for Tenders Document Schema is described as follows:

Document Schema Call for Tenders

Description: A document used by a Contracting Party to define a procurement project to buy goods, services, or works during a specified period.

Since the UBL is the basis for many e-Procurement systems, as described above, it is considered a well-established standard. Therefore, it cannot be neglected when developing the ePO. Especially the UBL concepts related to procurement, such as invitation for tenders, call for tenders, etc. should be carefully looked into.

In January 2016, the European Commission adopted the European Single Procurement Document (ESPD) , a document that aims to considerably reduce the administrative burden for companies, in particular SMEs who want to have a fair chance at winning a public contract.

To achieve this the ESPD maps out and replaces equivalent certificates issued by local public authorities or third parties involved in the procurement process, which can differ drastically between Member States.

While some countries have already introduced some form of “self-declaration” of suitability, others require all interested parties to provide full documentary evidence of their suitability, financial status and abilities. The ESPD will allow businesses to electronically self-declare that they meet the necessary regulatory criteria or commercial capability requirements, and only the winning company will need to submit all the documentation proving that it qualifies for the contract [_14].

To make full use of the ESPD concept, the European Commission will establish a service available for both suppliers and buyers, developing and providing the ESPD service free of charge to Member States and European Institutions. It will be provided as open source, so it can be implemented by service providers for their own use and to provide added value to buyers and suppliers [_15].

With regard to technical requirements, the transmission will be done through e-Tendering solutions. As the service works in conjunction with e-Certis, business registers and e-Tendering solutions, great care will be taken to harmonise the semantic data model. Development will be linked to e-SENS, the standardisation initiatives of CEN, the ISA Core Business Vocabulary, and solution providers.

In conclusion, the main objective of the ESPD is to reduce the administrative burden for buyers and suppliers participating in public procurement procedures. The ESPD service will reduce that burden by removing the need to produce a substantial number of certificates and documentation related to exclusion and selection criteria.

The ESPD initiative is worth examining carefully. As it maps all the certificates and evidence needed for procurement in the different Member States, it does the ePO a great service, as this is a task that will be necessary during the creation of the ePO.

e-Certis is a free online source of information to help companies and contracting authorities deal with the different forms of documentary evidence required in cross-border tenders for public contracts. e-Certis presents the different certificates frequently requested in procurement procedures across the EU [_18]. In particular, e-Certis can help companies to find out which certificates issued in their country they need to include in tender files submitted to an authority in any partner country, or contracting authorities to establish which documents issued by a partner country to confirm the eligibility of a tender are equivalent to the certificates they themselves require.

e-Certis is a reference tool and not a service of legal advice. The information contained in the database is provided by the national authorities and updated on a regular basis [_19].

e-Certis describes the documents using the following metadata:

e-Certis has a high reusability potential for our project as it could be a valuable reference when creating the classes and properties describing the certificates that are needed in the procurement process.

The ISA Core Vocabularies were created in collaboration with and by international working groups facilitated by the Interoperability Solutions for European Public Administrations (ISA) Programme of the European Union . Their aim is to facilitate the exchange of information in the context of European Public Services and address interoperability problems such as the lack of commonly agreed data models and universal reference data.

Core Vocabularies are simplified, re-usable and extensible data models that capture the fundamental characteristics of an entity in a context-neutral fashion. Public administrations can use and extend the Core Vocabularies in the following contexts [_20]:

Currently available vocabularies are:

Of the above vocabularies, the Core Criterion & Core Evidence, Core Business, Core Public Organization, and Core Person vocabularies can be especially useful for the eProcurement ontology as they describe fundamental parties and elements of public procurement contracts. Also, The Core Location vocabulary can provide a solution for describing any location data needed.

The PPROC ontology has been developed under the Public Procurement Optimization through Semantic Technologies Project (CONTSEM). This project is jointly undertaken by iASoft, the University of Zaragoza, ARAID (Government Agency of Aragon), the Government of Aragón, the Provincial Council of Huesca, and the town halls of Huesca and Zaragoza. The main purpose of the project is to add semantic technologies to the software used by public authorities for procurement procedures to publish data about public contracts. More specifically, one of the core objectives is to describe, semantically, the information published in official procurement bulletins [_21]. CONTSEM participants developed the PPROC ontology in accordance with Spanish laws and European laws in general.

The PPROC ontology defines the necessary concepts needed to describe the public procurement process and its contracts by effectively extending the Public Contracts Ontology. The main Class of the PPROC ontology is the pproc:Contract class, as the contract is considered to be the core concept of every procurement, as represented in Figure 3.

The other core Classes of the ontology which describe different aspects of procurement are the following represented in Figure 4: core classes of PPROC [21].

To describe all other concepts relevant to procurement contracts, the ontology reuses various ontologies and schemes. For example, the following solutions are reused: the Organization Ontology, the Schema.org scheme, the Simple Knowledge Organization System (SKOS) ontology, the Good Relations Ontology, and the Dublin Core Metadata terms scheme.

The PPROC ontology examined as a possibility for reuse in the ePO as it is extensive in its coverage, compatible with European procurement processes, well documented, and already extensively reuses existing established vocabularies.

LOTED (Linked Open Tenders Electronic Daily) is an ontology for the representation of European public procurement notices developed by the Knowledge Media Institute of the Open University.

It was created following the initiatives around the creation of linked data-compliant representations of information regarding tender notices in Europe, with the aim of addressing a specific problem plaguing previous efforts. Until now projects developing legal ontologies have attempted to represent the legal concepts and the case-based reasoning behind them in linked data simply by mapping them in OWL. However, due to the high level of detail and the nuances of legal reasoning, this approach resulted in extremely complex vocabularies. Complexity is unwanted in semantic applications because for intelligence to arise from linking heterogeneous data, the datasets in question must be flexible enough to integrate effectively.

The LOTED2 model seeks to find a balance between accurately representing the complex legal concepts and the reasoning behind them, and retaining the usability required for semantic applications. [_22]

Specifically LOTED2 has been designed for the following purposes:

LOTED2 is organized into the following 10 independent and reusable core modules which collectively represent 180 Classes:

In the case of ePO, the LOTED2 vocabulary could be useful as a means of enriching the data represented by the ePO with legal context. Also helpful is the fact that it is already designed with compatibility with TED data in mind.

The DNCP (National Public Procurement Portal) of open data, set up by the government, was created to provide access to data of public procurement in Paraguay and promote the development of creative tools that were attractiveof and service citizens. This initiative aimed to promote transparency, efficiency, citizen participation, and economic development by exposing the work done in various institutions, showing how they are managed and how they invest public resources. Table 12 in annex 7.2 lists all Classes used in the Paraguayan Procurement Ontology. Although the Paraguayan Procurement ontology aims to serve a similar purpose as the ePO, two problems with regards to its reuse were identified. First, the Paraguayan Procurement ontology is modelled completely in Spanish, which limits its reusability in the multilingual EU context. Secondly, the ontology is tailored to the local process. However the ideas behind the ontology could provide an interesting insight.

The Single Electronic Data Interchange Area (SEDIA) is a major strategic initiative that aims to create a master data repository of external stakeholders making business with the European Commission, whether business means grants or tenders.

The goal of the SEDIA project to create a fully automated and integrated process for handling procurement and grants information, strictly limiting the manual input of data to a minimum, and promoting the alignment and reuse of such data along the whole process. This requires the implementation of solutions based on interoperability of the different systems.

This is a process where the actors would not have to submit recurrent information over and over again, but would allow reuse of information previously submitted. Each piece of data that needs to be dealt with should be encoded only once, and then reused or updated according to the needs.

In order to achieve the envisaged interoperability a basic common understanding of the data dealt with is required. Therefore a common data model is to be created.

The SEDIA vocabulary is currently a work in progress. It started by mapping all relevant existing vocabularies and standards to ensure that it achieves its envisioned interoperability, and is in the process of creating a vocabulary.

In this vocabulary we describe all concepts that are part of the procurement process, and whose attributes are relatively static over time, as this is a vocabulary aiming to underpin a repository of stakeholders. Examples of such information are business and organization addresses, names, formal IDs, banking details etc.

The SEDIA vocabulary could be reused in the ePO to represent details about all kinds of stakeholders of the procurement process.

The Common Data Model (CDM) is the metadata model of the resources published by the Publications Office of the EU. The model is based on the FRBR model, being able to represent the relationships between the resource types managed by the Publications Office. Initially the focus was on metadata related to legal resources and general publications. in a later phase metadata for TED and CORDIS were added. The CDM includes different classes and properties that relate to e-Procurement . The CDM wiki explains which classes and properties are defined in the CDM and how they relate to each other. For example, the CDM defines a Public Procurement class as any of the works related to public procurement (Ted). The model also defines a Prior Information Notice class as a subclass of Public Procurement. The Public Procurement has, among others, the following properties:

Besides defining classes and properties, the CDM also defines relationships between concepts, such as:

The CDM can help us understand how different metadata concepts of e-Procurement relate to each other. The ePO will respect the naming and design rules of the CDM. Moreover, as the CDM is available in OWL, its elements can be reused by the e-Procurement Ontology wherever possible.

Following the adoption of the revised e-Procurement Directives, a new set of standard forms for public procurement was introduced. With the new directives, the forms are meant to be used in an electronic format only, which allows for automatic checking of mandatory fields. Moreover, the clear structure of electronic notices ensure consistency with the European Directives and minimise the risk of encoding errors. The forms, which are available via SIMAP, impose a structure for submitting the following information:

The standard forms for public procurement are very important for the development of the ePO as they describe how public procurement data should be submitted for publication in order to comply with the public procurement directives. Since the ePO has to be compliant with the same directives, it should take into account the concepts, data structure and controlled vocabularies of the standard forms for public procurement. Moreover, in 2015, the Publications Office and the ISA Programme of the EU conducted a study to elicit information and functional requirements from TED reusers [_23]. The requirements identified by this study could be considered when developing the ePO.

The Common Procurement Vocabulary (CPV) was created by the European Commission in order to facilitate the processing of invitations to tenders published in the Official Journal of the EU by means of a single classification system to describe the subject matter of public contracts. This classification endeavours to cover all requirements for supplies, works and services [_24].

The CPV consists of a main vocabulary for defining the subject of a contract, and a supplementary vocabulary for adding further qualitative information. The main vocabulary is based on a tree structure comprising of codes of up to nine digits associated with a wording that describes the supplies, works, or services forming the subject of the contract.

For example, if a contracting entity wants to obtain a road transport service for a fragile high-tech device, it may be interested in looking into the following codes:

Another example could be if an entity is interested in buying general-purpose rolling machines and parts for them. In order to find the most suitable codes, it could look into the following codes:

The supplementary vocabulary may be used to expand the description of the subject of a contract. The items are made up of an alphanumeric code with a corresponding wording allowing further details to be added regarding the specific nature or destination of the goods to be purchased.

For example, specific metals may be designated with the supplementary vocabulary codes: AA08-2 (Tin) or AA09-5 (Zinc).

The use of the CPV is mandatory for all public procurement procedures in the European Union as from 1 February 2006 [_25].

The CPV should be used in the case of the ePO as it is obligatory by directive. Furthermore as it is a wide spread and well established standard, its inclusion will facilitate integration and reuse of published data. An update of these CPVs are also foreseen within the ISA action: European Public Procurement Interoperability Initiative which also covers the ePO.

The Named Authority Lists (NALs) are harmonised code lists with multilingual labels used to facilitate data exchange. They are maintained by the Publications Office of the European Union in the Metadata Registry under the governance of the EU’s Interinstitutional Metadata Maintenance Committee (IMMC).

The use of common, high-quality reference data in information reuse can significantly reduce semantic interoperability conflicts. Available in different machine-readable formats and maintained by a trusted authority, the NALs can be reused in many different information exchange contexts.

Some examples of NALs that could be used in the domain of e-Procurement are those on countries, currencies, documentation types, EU programmes and EU corporate bodies [_26].

The Nomenclature of Territorial Units for Statistics (NUTS), is a geographical nomenclature subdividing the economic territory of the European Union into regions at three different levels: NUTS 1, 2 and 3 respectively, moving from larger to smaller territorial units, as it is shown in Figure 6.

The NUTS classification is a hierarchical system for dividing up the economic territory of the EU for the purpose of [_27]: * The collection, development and harmonisation of European regional statistics; * Socio-economic analyses of the regions; and NUTS 1: major socio-economic regions NUTS 2: basic regions for the application of regional policies NUTS 3: small regions for specific diagnoses * Framing of EU regional policies. Regions eligible for support from cohesion policy have been defined at NUTS 2 level. ** The Cohesion report has so far mainly been prepared at NUTS 2 level. Despite the aim of ensuring that regions of comparable size all appear at the same NUTS level, each level still contains regions which differ greatly in terms of population. Further details can be found the NUTS publication 2013/EU-28 .

The NUTS classification should be used for the purposes of the ePO when describing localities as it is obligatory in the public procurement directives .