Title: Adast: a decision support approach based on an ontology and CBR. Application to railroad accidents

Authors: Ahmed Maalel; Lassad Mejri; Henda Hajjami Ben Ghézala

Addresses: National School of Computer Sciences, ENSI, RIADI Laboratory, University of Manouba, Tunisia ' National School of Computer Sciences, ENSI, RIADI Laboratory, University of Manouba, Tunisia ' National School of Computer Sciences, ENSI, RIADI Laboratory, University of Manouba, Tunisia

Abstract: Recently, an increasing number of companies and industries have undergone greatly in competition. At the same time, we are witnessing an explosion technological advances and new technologies of information and communication that companies must integrate to achieve the performance that goes far beyond those obtained by conventional practices. However, these constraints are at the origin of the birth of many risks. Sometimes we are witnessing serious and costly failures, accidents and human losses, especially when it is a highly risky area such as railroad transportation (our current case study). This paper aims at developing a decision support approach, called Adast. The approach adopted in this research is based on acquiring and reusing past accident scenarii, historically validated on other homologated transport systems. It is composed of two main parts: knowledge models described by an ontology, and a reasoning process based on case-based reasoning (CBR). In this article, we present the architecture of the approach, the case model, the key processes, and the first steps of the experimental validation through the model feasibility based on Adast.

Keywords: decision support systems; DSS; artificial intelligence; case-based reasoning; CBR; ontology; accident scenarios; security; railroad accidents; railway accidents; rail transport; rail accidents.

DOI: 10.1504/IJIDS.2016.076507

International Journal of Information and Decision Sciences, 2016 Vol.8 No.2, pp.125 - 152

Received: 20 Jun 2014
Accepted: 06 Dec 2014

Published online: 11 May 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article