LIP6 2000/010

• Reports
  Modélisation de raisonnements tenus en contexte et application aux agents d'aide à la gestion d'incidents de SART
• L. Pasquier
• 45 pages - 03/21/2000- document en - http://www.lip6.fr/lip6/reports/2000/lip6.2000.010.ps.gz - 1,407 Ko
• Contact : Laurent.Pasquier (at) nulllip6.fr
• Ancien Thème : SYSDEF
Parisian subway is a complex dynamic system, where time plays an important role. When an incident occurs, the operators responsible for trafic regulation have few minutes to analyse the incident, its context and to plan an adapted solution for this situated incident. For one century, the parisian public transportation compagny recover this experience and establishes procedures. However, these procedures are only exemples for incident resolution, because they do not take the rich context into account. SART project (French accronym for Trafic Regulation Support System), and precisely the Incident Management Support Sub-System, will help the operators to construct their solutions. To facilitate the interactivity between operators and the system, we decide to base the artificial reasoning on human ones. As a consequence, we introduce the contextual graph paradigm, which appears as a computer solution to schemes that are used in psychology to describe human activities.
• Keywords : Decision Support System, context, contextual knoledge, proceduralised context, decision tree, contextual graphs, action schemes, case-based reasoning, industrial application
• Publisher : Valerie.Mangin (at) nulllip6.fr