Guideline for the dependability-oriented design of self-optimizing systems (bibtex)
by Christoph Sondermann-Wölke, Tobias Hemsel, Walter Sextro, Jürgen Gausemeier, Sebastian Pook
Abstract:
Self-optimizing systems are able to adapt their behavior autonomously according to their current self-determined objectives. Unforeseen influences could lead to dependability-critical behavior of the system. Methods are required which secure self-optimizing systems during operation. These methods to increase the dependability of the system should already be taken into consideration in the design process. This paper presents a guideline for the dependability-oriented design of self-optimizing systems, which integrates established classical methods like failure mode and effects analysis as well as methods based on self-optimization. On the one hand self-optimization is used to increase the dependability of the system by integrating objectives like safety, availability, and reliability to the objectives of the system. On the other hand methods are required to ensure the self-optimization itself. As basis for this guideline serves the principle solution of the system. The six phases of the guideline extend the design process and lead to an enhanced principle solution. Additionally, the guideline illustrates phases to implement and validate the self-optimizing system. The proposed guideline is applied to an innovative rail-bound vehicle, called RailCab, which is equipped with self-optimizing function modules.
Reference:
Sondermann-Wölke, C.; Hemsel, T.; Sextro, W.; Gausemeier, J.; Pook, S.: Guideline for the dependability-oriented design of self-optimizing systems. Industrial Informatics (INDIN), 2010 8th IEEE International Conference on, 2010.
Bibtex Entry:
@INPROCEEDINGS{Sondermann-Woelke2010b,
  author = {Sondermann-W{\"o}lke, Christoph and Hemsel, Tobias and Sextro, Walter
	and Gausemeier, J{\"u}rgen and Pook, Sebastian},
  title = {Guideline for the dependability-oriented design of self-optimizing
	systems},
  booktitle = {Industrial Informatics (INDIN), 2010 8th IEEE International Conference
	on},
  year = {2010},
  pages = {739 -744},
  month = july,
  abstract = {Self-optimizing systems are able to adapt their behavior autonomously
	according to their current self-determined objectives. Unforeseen
	influences could lead to dependability-critical behavior of the system.
	Methods are required which secure self-optimizing systems during
	operation. These methods to increase the dependability of the system
	should already be taken into consideration in the design process.
	This paper presents a guideline for the dependability-oriented design
	of self-optimizing systems, which integrates established classical
	methods like failure mode and effects analysis as well as methods
	based on self-optimization. On the one hand self-optimization is
	used to increase the dependability of the system by integrating objectives
	like safety, availability, and reliability to the objectives of the
	system. On the other hand methods are required to ensure the self-optimization
	itself. As basis for this guideline serves the principle solution
	of the system. The six phases of the guideline extend the design
	process and lead to an enhanced principle solution. Additionally,
	the guideline illustrates phases to implement and validate the self-optimizing
	system. The proposed guideline is applied to an innovative rail-bound
	vehicle, called RailCab, which is equipped with self-optimizing function
	modules.},
  bdsk-url-1 = {http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5549490},
  bdsk-url-2 = {http://dx.doi.org/10.1109/INDIN.2010.5549490},
  doi = {10.1109/INDIN.2010.5549490},
  file = {Sondermann-Woelke2010b.pdf:Sondermann-Woelke2010b.pdf:PDF},
  keywords = {RailCab;dependability-critical behavior;dependability-oriented design;failure
	mode;rail-bound vehicle;secure self-optimizing systems;self-optimizing
	function modules;optimisation;railways;self-adjusting systems;},
  timestamp = {2013.09.18},
  url = {http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5549490}
}