A Method Based on Bayesian Networksfor Estimation of Test Condence
Viking Flyhammar · KTH Publication Database DiVA (KTH Royal Institute of Technology) · 2015
In the reliability analysis community Bayesian Networks have gained increasedinterest and is a eld of research also in the automotive industry company ScaniaCV AB. At Scania, testing of requirements is widely used to ensure functionalityand safety for vehicles. Sometimes testing of a requirement does not exploreall of its possible behaviours and at Scania it is desired to guarantee that allrequired behaviors hold. To estimate the quotient of required behaviours thathold based on testing, a method using Bayesian Networks is proposed in thisthesis to estimate this quotient. It provides a novel framework for modellingrequirements in a hierarchical structure that calculates an estimate of the quotientof required behaviors that hold. Graphs created with the method visualizeswhich requirements that aects the quotient the most.1