Case Study

Abstract

This chapter illustrates the application of DST and MOEAs for quantifying and optimizing the uncertainty in a fault tree case study from the automotive field. The object of interest, an automatic transmission system under development emanates from the ZF AS Tronic family. The study represents the contribution of the University of Duisburg-Essen to the ESReDA (European Safety, Reliability & Data Association, [51]) workgroup on uncertainty analysis. This workgroup which unites industrial and research institutions (e. g. EADS, Électricité de France, JRC Ispra & Petten) aims at developing guidelines on uncertainty modeling in industrial practice based on a common framework developed from exemplary case studies. The chapter is structured as follows. After presenting the background of the study and introducing the IEC 61508 (section 7.1), a more detailed presentation on the system and the corresponding fault tree is given in section 7.2 and section 7.3. Section 7.4 introduces the IEC 61508 according to which safety requirements of the ATM are specified and shows how to integrate system and safety requirements in the ESReDA uncertainty analysis framework. Sections 7.5 and 7.6 describe how DST can help to check if the safety requirements are fulfilled. In sections 7.7 and 7.8, possible improvements regarding safety are screened using feature models and evolutionary algorithms. The obtained result set, containing several promising solutions is analyzed and could be postprocessed by the responsible decision-maker.