A survey on the benefits and drawbacks of AUTOSAR

The use of software reference architectures plays a fundamental role in software development, as it could bring several benefits such as providing means to design applications' software architectures with higher productivity and quality. However, many organizations still find scarce the existing empirical evidence about the benefits and drawbacks of software reference architectures. Organizations need such evidence to make informed decisions whether or not to adopt a software reference architecture for the development and maintenance of software applications. In this context, this paper aims to gather evidence on AUTOSAR, a mature and accepted software reference architecture for automotive applications used worldwide by more than 180 organizations. We designed and executed a web-based survey addressed to practitioners with experience in using AUTOSAR. We obtained 51 valid responses. The survey results indicate that the most popular benefits of AUTOSAR are standardization (88%), reuse (80%) and interoperability (51%) whereas its most important drawbacks are complexity (65%), initial investment (59%) and learning curve (51%). The respondents of the survey also gave directions to handle the major drawbacks of AUTOSAR, such as the need of a tool environment to improve its usability and handle its complexity, and the need of more stable releases of AUTOSAR to decrease the cost of migrating among versions.

[1]  Jan Bosch,et al.  Architecture for embedded open software ecosystems , 2014, J. Syst. Softw..

[2]  Simon Fürst AUTOSAR – A Worldwide Standard is on the Road . , 2009 .

[3]  Xavier Franch,et al.  Towards guidelines for building a business case and gathering evidence of software reference architectures in industry , 2014, Journal of Software Engineering Research and Development.

[4]  Jürgen Dingel,et al.  Model transformations for migrating legacy deployment models in the automotive industry , 2013, Software & Systems Modeling.

[5]  Muhammad Ali Babar,et al.  Applying empirical software engineering to software architecture: challenges and lessons learned , 2010, Empirical Software Engineering.

[6]  Miroslaw Staron,et al.  Evolution of Long-Term Industrial Meta-Models -- An Automotive Case Study of AUTOSAR , 2014, 2014 40th EUROMICRO Conference on Software Engineering and Advanced Applications.

[7]  Peter B. Seddon,et al.  Towards the improved treatment of generalization of knowledge claims in IS research: drawing general conclusions from samples , 2012, Eur. J. Inf. Syst..

[8]  Stefan Biffl,et al.  Practical Experiences in the Design and Conduct of Surveys in Empirical Software Engineering , 2003, ESERNET.

[9]  Xavier Franch,et al.  A survey on quality attributes in service-based systems , 2015, Software Quality Journal.

[10]  Springer Fachmedien Wiesbaden,et al.  AUTOSAR — The Worldwide Automotive Standard for E/E Systems , 2013 .

[11]  Claes Wohlin,et al.  Empirical Research Methods in Software Engineering , 2003, ESERNET.

[12]  Xavier Franch,et al.  Benefits and Drawbacks of Reference Architectures , 2013, ECSA.

[13]  Galib Krdzalic,et al.  Software Architecture Without Autosar , 2014 .

[14]  Per Runeson,et al.  Guidelines for conducting and reporting case study research in software engineering , 2009, Empirical Software Engineering.

[15]  Rob J. Kusters,et al.  Software Reference Architectures - Exploring Their Usage and Design in Practice , 2013, ECSA.

[16]  Paul W. P. J. Grefen,et al.  A framework for analysis and design of software reference architectures , 2012, Inf. Softw. Technol..

[17]  Springer Fachmedien Wiesbaden AUTOSAR has Become Mature and Accepted , 2013 .