Cluster-Based WSA for Decision Support Bayesian Systems: Case of Prognostic in Maintenance Management

Imen Ben Brahim, Sid-Ali Addouche, Abderrahman El Mhamedi, Younes Boujelbene

2018

Abstract

A knowledge representation and reasoning from data have produced many models. Probabilistic graphical models, specifically the Bayesian Network (BN) have proved its worth. It is considered to be a very useful tool for representing uncertain knowledge and decision-making support. This presupposes availability of knowledge problem in the conditional probabilities form. However, one is often in a critical situation because of data are insufficient, partially unavailable or heterogeneous. Developing methods and techniques to reconstruct the corpus of data needed for decision making, especially via BN is called the ”knowledge elicitation”. Several elicitation methods exist but they are not always applicable, too demanding in expert knowledge or presenting limits. The most generic and useful is the Weighted Sum Algorithm (WSA) but it presents two major issues concerning the compatible parental configuration. In the present paper, we discuss what the literature proposes for the first one, then we develop the solution for the second and validate it via a case of pump failure prognostic tool based on Bayesian support decision.

Paper Citation

in Harvard Style

Brahim I., Addouche S., Mhamedi A. and Boujelbene Y. (2018). Cluster-Based WSA for Decision Support Bayesian Systems: Case of Prognostic in Maintenance Management.In Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO, ISBN 978-989-758-321-6, pages 222-230. DOI: 10.5220/0006878502220230

in Bibtex Style

@conference{icinco18,
author={Imen Ben Brahim and Sid-Ali Addouche and Abderrahman El Mhamedi and Younes Boujelbene},
title={Cluster-Based WSA for Decision Support Bayesian Systems: Case of Prognostic in Maintenance Management},
booktitle={Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,},
year={2018},
pages={222-230},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006878502220230},
isbn={978-989-758-321-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,
TI - Cluster-Based WSA for Decision Support Bayesian Systems: Case of Prognostic in Maintenance Management
SN - 978-989-758-321-6
AU - Brahim I.
AU - Addouche S.
AU - Mhamedi A.
AU - Boujelbene Y.
PY - 2018
SP - 222
EP - 230
DO - 10.5220/0006878502220230