Applying case-based reasoning to the storing and assessment of software error-effect analysis in railway systems
DARRICAU ; HADJ-MABROUK
Type de document
COMMUNICATION AVEC ACTES INTERNATIONAL (ACTI)
Langue
anglais
Auteur
DARRICAU ; HADJ-MABROUK
Résumé / Abstract
THIS PAPER PRESENT A MOCK-UP OF A TOOL FOR STORING AND ASSESSING SOFTWARE ERROR EFFECT ANALYSIS (SEEA) FOR THE AUTOMATIC DEVICES SAFETY OF TERRESTRIAL GUIDED TRANSPORT SYSTEMS. SEEA IS AN INDUCTIVE PROCESS WHICH ATTEMPTS TO DETERMINE THE IMPACTS AND SEVERITY OF SOFTWARE FAILURES. THE PURPOSE OF OUR WORK IS TO EXPLOIT HISTORICAL SEEA, WHICH HAVE BEEN ALREADY CARRIED OUT ON APPROUVED SAFETY-CRITICAL SOFTWARE, IN ORDER TO ASSESS SEEA OF A NEW SOFTWARE. THE PRODUCTION OF THIS MOCK-UP, IN THE PROCESS OF VALIDATION, INVOLVES THE USE OF CASE-BASED REASONING (CBR). THIS IS ONE OF THE REASONING TYPES USED IN ARTIFICIAL INTELLIGENCE FOR MACHINE LEARNING. THE BASIC PRINCIPLE OF CBR IS TO DEAL WITH A NEW PROBLEM BY REMEMBERING SIMILAR EXPERIENCES WHICH HAVE OCCURED IN THE PAST.
