Critical Observations in a Diagnostic Problem

James Christopher Cody 1 Marie-Odile Cordier 2 Alban Grastien 1
2 DREAM - Diagnosing, Recommending Actions and Modelling
Inria Rennes – Bretagne Atlantique , IRISA-D7 - GESTION DES DONNÉES ET DE LA CONNAISSANCE
Abstract : — We claim that presenting a human operator in charge of repairing a faulty system with a small subset of observations relevant to the failure improves awareness and confidence of the operator. Consequently, we introduce the problem of finding a set of relevant observations (called the critical observations) that can be used to derive the same diagnosis as the full problem. We show how this problem can be solved and illustrate its benefits on a real diagnostic problem.
Document type :
Conference papers
Complete list of metadatas

Cited literature [8 references]  Display  Hide  Download

https://hal.inria.fr/hal-01100627
Contributor : Marie-Odile Cordier <>
Submitted on : Wednesday, January 7, 2015 - 12:16:07 PM
Last modification on : Tuesday, September 17, 2019 - 5:08:02 AM
Long-term archiving on : Friday, September 11, 2015 - 12:46:37 AM

File

cdc14.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01100627, version 1

Citation

James Christopher Cody, Marie-Odile Cordier, Alban Grastien. Critical Observations in a Diagnostic Problem. IEEE Conference on Decision and Control (CDC-14), Dec 2014, Los Angeles, United States. ⟨hal-01100627⟩

Share

Metrics

Record views

428

Files downloads

174