Risk Management for Embedded Systems

Thomas Kurz
Salzburg University of Applied Sciences and Technologies

Susanne Hartkopf
Fraunhofer Institute for Experimental Software Engineering

Dr. Thomas J. Heistracher
Salzburg University of Applied Sciences and Technologies

Concept:

For software and manufactured products, a number of risk management methods exist. Firstly, this work presents an adaptation of a risk management method for manufactured products, the Failure Mode and Effects Analysis (FMEA), to software needs. Secondly, the adapted method, Software Mode and Effects Analysis (SREA), is compared in a case study with a risk management method especially designed for software development. A merged method is required that takes into account the originalities of software and manufactured products. This method should have the potential of saving costs, should make the system more reliable, and should enable faster development of software in embedded systems. It should provide an enhanced method that’s original, (FMEA), is well-known to customers and co-workers. This paper presents the results of a case study in which a software engineering risk management method (Riskit) and the adapted production process oriented risk management method (SREA), are applied to a mobile Microsoft .NET application for the configuration of industrial plants. Along with the verification of applicability of the new SREA approach, the results of the case study provide substantial considerable information for risk management in embedded systems.

Outline of the presentation:

• Risk Management at a Glance
• Limitations of current Approaches
• Software Risk and Effects Analysis (SREA)
• Empirical Evaluation
• Conclusions and Forecast

Learning Objectives:

• Learn, how a risk management method for embedded systems could be applied to a traditional engineering field
• Learn, why manufacturing risk management methods fail in the application for software products
• Learn, where the limitations of applying a software risk management method to systems with strong dependencies of hardware and software are
• Learn, which results were derived by applying a risk management method with special focus on interfaces between hardware and software
• Learn, what should be considered for further experimentation on risk management for embedded systems

Biography Dipl.-Ing.(FH) Thomas Kurz

Dipl.Ing.(FH) Thomas Kurz recently received his Master’s degree (Dipl.-Ing.(FH)) in Telecommunications Engineering in July 2003 and works currently as researcher in the field of Software Engineering in Salzburg, Austria.

He wrote his diploma thesis, named “Risk Management for Embedded Systems: The Software Risk and Effects Analysis (SREA)” at the Salzburg University of Applied Sciences and Technologies, Department of Informatics and Software Engineering in cooperation with the Fraunhofer Institute for Experimental Software Engineering, Germany. A joint case study was performed with a company in Salzburg, Austria. Within this case study, a software engineering risk management method, Riskit, and the adapted production process oriented risk management method, Software Risk and Effects Analysis (SREA), were applied to a mobile Microsoft .NET application for the configuration of industrial plants.

Biography Dipl.Inform. Susanne Hartkopf

Dipl.Inform. Susanne Hartkopf is a consultant and researcher at the Fraunhofer Institute for Experimental Software Engineering, Kaiserslautern, Germany.

Her experiences with software risk management are manifold. She has accompanied software projects in different sectors with risk management. Furthermore she holds regularly workshops about this topic. Finally, she currently performs her PhD in this area. She holds a MSc degree in computer science and economics.

Biography Prof. Dipl.-Ing. Dr. Thomas J. Heistracher

Dr. Heistracher is head of dept. of Informatics and Software Engineering at Salzburg University of Applied Sciences and Technologies. His research interests cover topics such as software engineering methodologies, spontaneous networking environments, adaptive middleware-systems, software agent systems, metadata usage in distributed computing, computational fluid dynamics and modelling of ecosystems.

Dr. Heistracher holds a Master’s degree in Electrial Engineering and Information Technolog y (Dipl.-Ing.), a PhD in Physics and Biophysics (Dr.rer.nat.) and is Professor since 1999.