Integrating model-based diagnosis techniques into current work processes --- three case studies from the INDIA project

  • Authors:

  • Heiko Milde;Thomas Guckenbiehl;Andreas Malik;Bernd Neumann;Peter Struss

  • Affiliations:

  • Laboratory for Artificial Intelligence, University of Hamburg, Vogt-Koelln-Str. 30, 22527 Hamburg, Germany E-mail: {milde, neumann}@informatik.uni-hamburg.de (Corresponding author);Fraunhofer-Institut IITB, Fraunhoferstr. 1, 76131 Karlsruhe, Germany E-mail: guc@iitb.fhg.de;ESG Elektroniksystem-und Logistik-GmbH, Germany E-mail: malik@esg-gmbh.de (Also affiliated with Robert Bosch GmbH during the project);Laboratory for Artificial Intelligence, University of Hamburg, Vogt-Koelln-Str. 30, 22527 Hamburg, Germany E-mail: {milde, neumann}@informatik. uni-hamburg.de;Technical University of Munich, Department of Computer Science, Orleansstr. 34, 81667 Munich, Germany E-mail: struss@in.tum.de

  • Venue:
  • AI Communications
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

Although the area of model-based diagnosis has developed a numberof prototypes with impressive features that promised economicimpact and, hence, caught industrial interest, the number of actualindustrial applications is still close to zero. One of the reasonsis that the successful techniques have not yet been turned intotools that reflect and support the current diagnostic workprocesses and their existing tools. The INDIA project joined eightGerman partners (research groups, software suppliers, and endusers) in an attempt to take a major step in the transfer ofmodel-based diagnosis techniques into industrial applications. Thispaper describes part of the work carried out in this project.Rather than presenting the theoretical foundations of thetechniques in depth, we focus on the aspect of how model-baseddiagnostic techniques can be related to established tools andsystems in order to provide some leverage for today's workprocesses and to change them gradually, as opposed to postulating aradical change in current practice and organizational structures.From this perspective, we discuss the utilization of model-basedtechniques for the generation of fault trees for on-line testingand diagnosis of fork lifters, generation of test plans for anintelligent authoring system for car diagnosis manuals, and theexploitation of existing state-chart process descriptions forpost-mortem diagnosis of processes in a dyeing plant.