Towards a new Information Theory & its Applications

Scientific stakes

The idea of ‘system’ is pervasive in our descriptions and in our projects. The impossibility to come up with a generally and scientifically valid definition of this idea touches the most profound features of the classical conceptualization, those that underlie the current languages, the grammars and the classical logic. These features are rigidly tied with the belief that what we know is what ‘really’ exists, such as we know it. But this belief is refuted by microphysics, where it appears that that we create our knowledge entirely, as much the ‘entities’ that we examine as the qualifications of these entities. The Method of Relativized Conceptualization (MRC) of Pr. Mugur-Schächter establishes that the processes of creation of knowledge are subjected to relativization, and that making them explicit clarifies the involved significances and permit to dominate the finalities. It is surprising to perceive to what a point the relativization required by the method winds up the ambiguities commonly tied up to the idea of system.

But such a clarification, to be effective, requires that this constructive handling of our conceptualization processes be turned into a formal system of general application. It entails a brand new Information Theory which brings into light the relativization acting in the conceptual-factual genesis of ‘classical’ basic concepts such as persistence, continuity, state or causality. Such a refoundation encompasses a physical relativized interpretation of the Measurement Theory, up to now purely mathematical, and as a consequence, of the concepts of law of probability, of entropy and of complexity. The work already achieved through an Infra-Quantum Mechanics (IQM), the Method of Relativized Conceptualization (MRC) and a Second Quantum Mechanics (MQ2) of Pr. Mugur-Schächter, and a Relativized Systemic (RS) of H. Boulouet paves the way.

The conceptual, tooled an operational developments already conducted, based on RS algebra, with a Relativized System Engineering – RSE (H. Boulouet) and a Knowledge Genesis Management – RIM (F. Fleuchey) have demonstrated the tremendous benefits which could be expected from such a fundamental upheaval.

Applications

RIM and RSE are two conceptual developments of MRC-RS. They gave rise to different prototypes of Integrated Development Environment (EDI) from 2005 to 2011.
RIM and RSE, together considered, provide an unprecedented relativized framework, finality driven, to master complex system design or knowledge building processes. This framework has already been successfully evaluated when applied to mechatronic systems design in the automotive industry.

Relativized System Engineering (RSE)

Relativized System Engineering (RSE) focuses on the formalization of concept genesis.
It provides a formal framework for the expression in a relativized and scientific way (refutability and repeatability), of the expected and of the realized. It covers the whole range of artefacts going from requirements to automated testing.

The aims are :

• To regard the development process as overlapping contributions, always formally distinguished, put in coherence on a to-be-designed material support, emerging from these different stakes and from accumulated knowledge and available means.
• To master the building process of consensual relationships between biopsychic perceptions (human usages) and technical descriptions (industrial standardization and repeatability).
• To ensure the testability of any model element at any abstraction level and to make possible automated evaluation (on test bench) of the realized against the expected.

To meet these expectations, RSE:

• Clearly distinguishes the biopsychic and technical-scientific models of some physical reality and specify in both cases the testing referential and the constructed dependencies between these two points of view.
• Organizes the development process as the confrontation of two antagonistic ways of thinking: focalization on a given finality which leads to the concepts of usage and system, on one hand, and on the other, the concern for technical optimization, which leads to the concepts of shared resource and architecture.
• Formalizes any physical item (entity, system) as the conceptual invariant which emerges from descriptions of expected or recorded dynamic processes, relativized by acting finalities and associated views.
• Makes possible to consider the different contributing points of view altogether or separately at any stage of development process.
• Standardizes conceptualization levels, built upon what is adopted as “elementary facts” : the transferred or basic description level (MRC concept).
• Doesn’t impose any task scheduling, but give explicit criteria to assess whether the model is consistent, refutable and meet its goals.

Relativized Information Management (RIM)

Relativized Information Management (RIM) focuses on the management of concept genesis.
From an industrial point of view, it provides a formal framework to define and manage available means, project organizations, tasks and their products, all that in a systematically relativized and consistent way.

Each contributor to a product development, according to his role, to a chosen point of view and to local stakes, participates in :

• Setting up and implementing the project organization – human and material resources –, processes.
• Goals definition, assignment and assessment.
• Project monitoring.
• The definition and monitoring of representational artefacts lifecycles (achievement status, applicability, effectivity …).