Technologies for Information Management
Please note that the project factsheet will no longer be updated. All information relevant to the project can be found on the CORDIS factsheet. This is updated on a regular basis.
ONTORULE - ONTOlogies meet Business RULEs
Ontologies meet business rules
The objective of ONTORULE is to enable the right people to interact in their own way with the right part of their business application: that means different people with different tasks, requirements, and objectives and with different background, knowledge and languages, ranging from business executives over business analysts to IT developers, who all interact in different ways with different aspects of a business application, to use, control, maintain and/or evolve it.
We believe that one essential step towards achieving that objective is the ability to separate cleanly the domain ontology from the actual business rules, on the one hand; and the representation of the knowledge from its operationalisation in IT applications, on the other hand. That is, the ability to separate the conceptual and structural knowledge from '[those] statements that define or constrain some aspect of the business [and that are intended] to control or influence the behaviour of the business', on the one hand; and the rules and ontologies from the data models and the executable code, on the other hand.
Leading vendors of knowledge-based and business rules management systems and a handful of top research institutions join their efforts, in ONTORULE, to develop the technology that will empower business professionals in the enterprise of the future.
Two large companies are the test beds that will ensure the success and business impact of the technology.
Suppose that you have to write the business rules that represent some policy in a rule-based application: for instance, your company’s pricing policy. Would you prefer a language that is based directly on the data model of the underlying application? Or would you rather write the rules in your own, or your company’s, vocabulary, based on the concepts you are used to in your everyday business life?
If you are the IT person in charge of the underlying application, you are likely to prefer the former; but you are unlikely to be the owner of your company’s pricing policy. If you are the business owner of the policy, we believe that you would prefer the latter, and our market research tells us that this is, indeed, what the business user wants.
The obvious key step forwards seems to be that the conceptual knowledge in which the vocabulary is grounded – and in which, as a consequence, the business rules are grounded as well – must be represented and maintained separately from both the IT application’s data model and the business rules themselves. Obvious? But, then, why has this not been done before?
One reason is that, in most cases, it is far from trivial for a business user to separate the terms and facts that express the ontology, from the business rules properly. And it would be farfetched, of course, that a business user would be willing, or able, to model the underlying formal ontology. So that the whole objective depends on developing and integrating the technology and the methodology that will be required to support the acquisition of the business vocabularies and underlying ontologies from all the available sources, including policy and regulatory texts, and the expression of the business rules themselves.
But, even with the right modelling technology, adapting authoring tools for business rules such that they refer to an ontology instead of a data model is not enough to allow the business user to write rules using his own language, grounded in his own ontology, where rules and ontologies are maintained separately.
Indeed, the owner of an item should never be required to evaluate, identify, or even be aware of the impact that modifications she makes might have on other items that are not under her management or ownership; not to mention implementing any changes required on such items as a result. In a similar way, the owner of a resource should not be expected to understand or answer questions regarding the correctness, consistency or validity of items that are not under her management or ownership, even if they need be traversed when debugging, validating or maintaining the consistency of resources she manages.
ONTORULE will have to solve technical and theoretical difficulties ranging from dependency management to consistency checking, debugging and validation, to issues related to scalability, accessibility, etc; and to develop and integrate the different technologies that will be required to make the ownership, management and maintenance of the interdependent rules and ontologies easy and usable for the business owner of each separate resource, in particular by making the dependency on other resources transparent to him.
Eventually, the rules must be implemented in the IT applications, and, however desirable it might be from the business user’s viewpoint, breaking the code-as-infrastructure model results in the business rules not being immediately implementable anymore: the ontologies must be mapped onto the data model on which the rules will operate in the IT application (e.g. an object model, a relational schema), the rules must be operationalised accordingly, and the whole process must be transparent for the business user.
The operationalisation of the business rules raises another set of theoretical, technical and practical problems that will need be resolved: typical ontology languages, such as OWL or SBVR, work on an open world assumption whereas implemented rule languages work usually on the opposite closed world assumption; the ontology will usually not map completely onto the IT data model; different families of languages are used, that have different semantics; not to mention efficiency issues, scalability etc. Ontology reasoning and rule-based inference will have to be integrated theoretically, and they will have to be technically combined engine-wise as well.
The objective of ONTORULE is to integrate all the pieces of knowledge and technology that are needed to make the ONTORULE vision happen, including some that will need be researched and developed within the project. For an integration that is both effective and flexible, ONTORULE will rely heavily on open standards: identifying and contributing to the specification of the appropriate standards is of essential importance to the project.
- Project Overview (2MB)
- Flyer (487KB)
- Annual Report 2010 (333KB)
- Annual Report 2009 (559KB)
- Project website
- Project coordinator
Christian de Sainte Marie, ILOG, France
Email to the ONTORULE coordinator
- ILOG, France (coordinator)
Ontoprise GmbH Intelligente Loesungen fuer das Wissensmanagement, Germany
Libera Universita di Bolzano, Italy
Technische Universitaet Wien, Austria
PNA Training BV, Netherlands
Iniversite Paris 13, France
Fundacion Centro Tecnologico para el Desarrolo en Asturias de las Tecnologias de la Informacion, Spain
Audi Aktiengesellschaft, Germany
Arcelormittal Espana SA, Spain
- Administrative details
- ONTORULE (ICT-231875) is an Integrated Project (IP) of the European Union's 7th Framework Programme: Information and Communication Technologies (ICT) – Call 3.
The project started on 1 January 2009, and will finish on 31 December 2012 (36 months).
There are 9 participants from 6 countries involved in the project, and the EC contribution is 5.4 million Euros (total cost: €8.03m).
This page is maintained by: INFSO.E2 Webmaster