The aim of this paper is to provide a conceptual and methodological framework in order to prevent knowledge loss in a long duration space project.
Starting from risk management, the paper considers existing factors that contribute to the success of the mission, such as dependability and safety, and then argues, using a multi‐viewpoint approach, that risk analysis produces knowledge (not simply information or data). Then, the paper describes how the filtering of risky components of a technical documentation is performed. It is based on the confrontation of the vocabulary of the different documents to an ontology of “criticality” built by the authors. The paper also describes how the knowledge evolutions are detected and how the interpretation of these evolutions is carried out.
On a conceptual side, a general model of the design process is presented based on a multi‐viewpoints approach and characterised by a value system. On the practical side, an ontology of risk, used as a reference system in order to compare knowledge at different stages of a project, is described.
Some difficulty arises when a very huge documentation is addressed. Among all evolution clues a lot of them could be well‐known by everybody or could be of little importance.
The paper intends to have a preventive strategy for knowledge loss in a long duration project. Reaching the ultimate stage of a mission, project management should be able to identify the main knowledge differences between technical culture of new incomers and the one of the early designer that can be found in the project documents.
The paper carries a multi‐discipline approach, putting together different domains: space activity, statistic specialist, knowledge managements, and linguistics.
