Set up in 1965 under initiative of mechanical companies and their professional organisation, the purpose of CETIM is to conduct studies and research for the French mechanical industry. Its extended scope of competencies covers the various requirements of that industry in terms of design, production, maintenance, control and measurement, environment administration and management. CETIM has a strong awarness of the need of engineering industry. In close cooperation with CETIM, the Federation of the mechanical engineering industries (FIM) seized the appropriateness of the publication in 2006, by the French Ministry for Industry, the study "key technologies for industry by 2010". CETIM is a key partner of the FP6 Network of Excellence, I*PROMS and is involved in 300 on going Research & Development projects including 12 new European contracts in 2006.

INRIA (National Institute for Research in Computer Science and Control) is a French public-sector scientific and technological institute operating under the dual authority of the Ministry of Research and the Ministry of Industry. INRIA missions are "to undertake basic and applied research, to design experimental systems, to ensure technology and knowledge transfer, to organise international scientific exchanges, to carry out scientific assessments, and to contribute to standardisation".

The research carried out at INRIA brings together experts from the fields of computer science and applied mathematics covering the following areas: Networks and Systems; Software Engineering and Symbolic Computing; Man-Machine Interaction; Image Processing, Data Management, Knowledge Systems; Simulation and Optimisation of Complex Systems.

INRIA gathers in its premises around 2 100 persons including 1 600 scientists, many of which belong to partner organisations (CNRS, industrial labs, universities) and are assigned to work in common "projects". On INRIA budget, around 500 full-time equivalent R&D positions can be accounted for. A large number of INRIA senior researchers are involved in teaching and their PhD students (about 550) prepare their thesis within the different INRIA research projects (currently 74).

ROBOSOFT established in 1985 by INRIA researchers, is recognized as a leading supplier of advanced robotics solutions throughout Europe and beyond. ROBOSOFT is a profitable and independant sme of 20 persons with a turnover of 3 M€. ROBOSOFT has been building a unique expertise in service activities automation. Its experience in robotics, more precisely in subsets integration and automatic control, allows to provide operational robotic solutions in 5 areas :

• Education and research in robotics
• Transport : automatic transport of goods and people, automatic gas refuelling
• Cleanliness : automatic floor and windows cleaning
• Security : remotely operated mobile robots
• Health : assistive mobile robots for elderly and handicapped people

While education, transport, cleanliness and security have reached a commercial level with identified products and running applications, health area is still at the research level. ROBOSOFT has gathered, within its team of engineers, the skills needed for a perfect knowledge of techniques linked with robotics: real-time and on-board computing (hardware and software), mechanichal design, electronics, automatic control, prototype ande small serie manufacturing … Willing to widen its range of products, ROBOSOFT developed a distribution activity and offers now a large choice of components and generic subsets for developers of advanced robotics solutions.

The Research Laboratory "Génie de Production" belongs to the National Engineering School of Tarbes, France. It is recognised by the French CNRS as an "équipe d'accueil" (EA n°1905) and is composed of 35 senior researchers, 25 PhD students and 6 technical staff divided into three teams: Material Engineering, Mechanical Engineering and Computer Integrated Manufacturing.

The Computer Integrated Manufacturing team groups 15 senior researchers and 14 PhD students and leads several research projects relevant to the "I*PROMS " themes. The common objective of these studies is to improve the productivity, quality, flexibility and reactivity of the manufacturing processes and systems by a better use of information and knowledge. Two main types of methods are considered in that purpose: analytical control methods when models of the processes are available, and artificial-intelligence based methods when expert knowledge regarding the system is available. The CIM team has especially conducted researches in the following areas:

Model-based monitoring and control accommodation of Machine Tools

The machining process is a complex phenomenon which depends on the material, on the cutting operation, on the machine and on the cutting tool. An efficiently controlled system should be able to use information regarding various these various aspects. Techniques have been developed for an efficient on-line monitoring systems, using a cutting process model integrating tool wear, flexion and chatter in High Speed Machining (a PhD is in progress on this subject with Turbomeca).

Mobile robots for reconfigurable manufacturing systems

For increasing the flexibility of manufacturing systems, it will shortly be required to be able to reconfigure a workshop according to a short term demand. In that purpose, the team is performing research on mobile robots for manufacturing activities (ROBEA I and II projects, supported by the French CNRS).

Distributed decision making for intelligent machines supervision

Communication and decision capabilities can be added to the functional layer provided by smart sensors and actuators by defining cooperation scenarios. The multi-agent paradigm is an efficient way to implement empirical operational knowledge when accurate models of the process are not available: a global architecture for cooperation of smart sensors and actuators has so been suggested, implemented and tested in a multi-agent environment.

Management of the demand uncertainty in Supply Chains

The customers demand being more and more uncertain, it is important that production management techniques allow to explicitly deal with this uncertainty all along the Supply Chain. A method has been suggested for allowing to take into account expert knowledge for modelling this uncertainty thanks to fuzzy logic, then to propagate this uncertainty all along the planning levels. This should allow the decision makers to have a better image of the various situations that may occur in the near future, on the points of views of material requirement planning and possible load (project PICHALOG, supported by the CNRS).

Knowledge Engineering for manufacturing Systems

The major point in order to improve the behaviour of manufacturing systems is to be able to identify, structure, store and re-use the knowledge available in the companies. In that purpose, several studies have been launched in companies on subjects like Case-based reasoning, return on experiment or identification of human resource's roles, competence and knowledge in decisional processes (PhD with Alstom).