Ph. D. Project
Formal methods for business knowledge extraction in cooperative system engineering
2015/10/01 - 2018/09/30
One approach to the engineering of interoperating systems is to be based on different types and different levels of abstraction or
models. These models must express not only the aspect of "structural" component of the system, but also their behavior, the latter
being constrained by domain-specific requirements of the system (business rules). Another type of constraint can be induced by
interoperation protocols that can impose strict rules to equip interoperating systems with properties such as autonomy,
confidentiality and transparency.
The aim of this work is to study the problems of cooperative systems engineering models. To remain competitive, more and more
companies are forced to work together, opportunistically or stabilized in a sustainable way or ephemeral, with other companies
with the aim to optimize costs and production time, the quality and marketing of their products and/or services (Thoben and Jagdev
These collaborative enterprises are organized now, as part of business networks whether in the form of extended enterprise or
virtual firms (Bititci, et al. 2004) (Camarinha-Matos and Afsarmanesh 2008).
Defined as such, these collaborative companies can be assimilated to a reticular system of open enterprise systems (Oberndorf
1998), and by extension the information system (IS) of the network itself can be considered a reticular system of SI. The
specification of such a network of SI means moving from one integration paradigm to a paradigm of interoperation (Fisher 2006).
One of the requirements of this need for collaboration focuses on the ability of these components to interoperate, that is to say,
their interoperability, more or less complete. Ducq (Ducq 2008) considers the interoperability of systems as a specific performance
requirement of the company.
There are standard and reference tools providing practical and metrics to measure interoperability. For example, we can mention,
without being exhaustive, EIF (European Interoperability Framework) LISI (Level of Information Systems Interoperability) LCIM
(Levels of Conceptual Interoperability), FCA (Formal Concept Analysis) (in French Formal Concepts Analysis).

Various research studies have proposed maturity models and formal metrics to assess the potential or the degree of semantic
interoperability companies wishing to establish a collaborative network. However, these results do not allow full automation of the
evaluation process because they suffer from a computational formalization of their models.

The scientific challenge is how to provide languages and modeling tools for each business modeling project, despite the
heterogeneity of professional skills and multidisciplinary areas. This challenge has two dimensions: first, that of the model's ability
to equip business processes, which requires the definition and formalization of their invariants; secondly, the study of the
conditions of use of the models in practice, always evolving and uncertain, the company's business.

The formal concept analysis is a useful and powerful tool to formally describe the links between any objects (which form a context),
particularly between objects conveying knowledge. This method is based on lattice theory, which can be used to resolve
interoperability issues between evaluation of information systems within companies.

An extension of the CFA mechanisms was introduced (Rouane et al. 20013) and called Relational Concepts Analysis (RCA), where
the focus is on the sets of data compatible with the Model Entity Relationship (ER) or, alternatively, with RDF (Resource Description
Framework). This is a method to extract the conceptual knowledge from multi-relational data.

Faced with this challenge the identified scientific problems are:
• the lack of formalization (also mathematical) of the system models and of the information systems. The definition of the
semantics of concepts and relationships that is used to ensure common understanding and facilitates the semantic interoperation
minimizing the semantic losses;
• the use of algebraic and / or geometry tools (lattice theory, category theory, homological algebra) in the context of formal
concept analysis, which is a recent approach that has not yet been fully developed (even the mathematical point of view);

This thesis, involving both theoretical and applied, is proposed as part of the Research Federation "Charles Hermite". Its
theme is within the research areas "Interoperability of distributed heterogeneous systems" and "Structures and basic tools" of the
federation. It fits complementarily in the SIA project research themes (Ambient Intelligent Systems) of the thematic group ISET
CRAN and the IECL Geometry team. Internationally, this is partially processed in the European virtual laboratory INTEROP-VLab
Interoperability whose members "cover" a large part of Europe and China. The outcome of research community IFAC TC
5.3 "Enterprise Integration and Networking" is also interested in the issue of formalizing the semantics and models for systems

In this thesis it will be necessary to acquire knowledge of the lattice theory, particularly lattice concepts, in order to master the
analysis of formal concepts. In relation to the direction that the study of the issue addressed will may be considered that the
student acquire further expertise in the field of algebra.
Interoperable systems engineering, models, interoperability, formal concept analysis
The PhD thesis will last three years.
Eco-Technic systems engineering