Skip to Main content Skip to Navigation
Journal articles

Models and algorithms for failure diagnosis and recovery in FMS

Armand Toguyeni 1 Pascal Berruet 2 Etienne Craye 1
2 Lab-STICC_UBS_CACS_MOCS
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance
Abstract : Within the FMS framework, supervision has been introduced to take into account dependability objectives in FMS control. This paper focuses on two modeling lines used to implement some functions of FMS monitoring and supervision: a functional line and a structural line. The model obtained by the first line is called a functional graph (FG). The FG provides a graphic description of logical conditions to assume services delivered by the plant from functions of its components. It is used in diagnostic and prognostic procedures. The structural line is called operational accessibility graph (OAG), which models the operations of an FMS, taking characteristic areas and accessibility relationships into account. It represents the way operations are induced to cooperate. It is used in a recovery procedure to establish the FMS configuration and the operations' sequences in order to achieve a required production. From a modeling and application point of view, FG and OAG are complementary. They are used to implement a reaction loop of supervision that starts from a failure's detection to the proposal of a new FMS organization. In order to improve the dependability of an FMS during exploitation, a cooperation of the two models is proposed.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-00080207
Contributor : Pascal Berruet <>
Submitted on : Thursday, June 15, 2006 - 11:02:16 AM
Last modification on : Tuesday, November 24, 2020 - 2:18:22 PM

Identifiers

  • HAL Id : hal-00080207, version 1

Citation

Armand Toguyeni, Pascal Berruet, Etienne Craye. Models and algorithms for failure diagnosis and recovery in FMS. International Journal of Flexible Manufacturing Systems, Springer Verlag, 2003, Vol 15, N°1, pp.57-85. ⟨hal-00080207⟩

Share

Metrics

Record views

321