• DocumentCode
    1637394
  • Title

    Software reliability allocation with safety concerns in medical devices

  • Author

    Chen, Qi ; Wei, Yuan ; Qin, Jin

  • Author_Institution
    Univ. of Sci. & Technol. of China, Hefei, China
  • fYear
    2012
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    Software reliability allocation is a complicated multiple-criteria decision process which involves many factors such as cost, performance, functional complexity and testing. For medical devices, safety is an extremely critical factor that it must be emphasized in the system software development from the early stages. In this paper, we demonstrate an intuitive approach to incorporate safety into reliability allocation. This approach starts with allocating reliability to software modules associated with specific safety feature through functional decomposition and fault tree analysis, and finalizes the allocation for each software module by considering all safety features. This approach can provide a conservative and practical safety-related reliability allocation in an easy-to-use way at software planning and design stages. The proposed approach is shown in detail with a simplified real case.
  • Keywords
    biomedical equipment; fault trees; medical computing; object-oriented methods; safety-critical software; cost factor; fault tree analysis; functional complexity factor; functional decomposition; medical devices; multiple-criteria decision process; performance factor; safety concerns; safety feature; safety-related reliability allocation; software design stage; software planning stage; software reliability allocation; system software development; testing factor; Educational institutions; Fault trees; Resource management; Safety; Software; Software reliability; fault tree analysis; safety; software reliability allocation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Reliability and Maintainability Symposium (RAMS), 2012 Proceedings - Annual
  • Conference_Location
    Reno, NV
  • ISSN
    0149-144X
  • Print_ISBN
    978-1-4577-1849-6
  • Type

    conf

  • DOI
    10.1109/RAMS.2012.6175478
  • Filename
    6175478