Metric Suite for Directing the Failure Mode Analysis of Embedded Software Systems.pdfVIP

METRIC SUITE FOR DIRECTING THE FAILURE MODE ANALYSIS OF EMBEDDED SOFTWARE SYSTEMS Guido Menkhaus, Brigitte Andrich Computer Science Department University of Salzburg, Austria guido.menkhaus@cs.uni-salzburg.at, bandrich@cosy.sbg.ac.at Keywords: Failure mode and effects analysis, software metrics, software design Abstract: Studies have found that reworking defective requirements, design, and code typically consumes up to 50 percent of the total cost of software development. A defect has a high impact when it has been inserted in the design and is only detected in a later phase of a project. This increases project cost, time and may even jeopardize the success of a project. More time needs to be spent on analysis of the design of the project. When analysis techniques are applied on the design of a software system, the primary objective is to anticipate potential scenarios of failure in the system. The detection of defects that may cause failures and the correction is more cost effective in the early phases of the software lifecycle, whereas testing starts late and defects found during testing may require massive rework. In this article, we present a metric suite that guides the analysis during the risk assessment of failure modes. The computation of the metric suite bases on Simulink models. We provide tool support for this activity. 1 INTRODUCTION and thus the detection of the number of risky compo- nents of a system. This can lead to a misunderstand-