IV&V Facility Research Program Results and SARP Results  

Advanced Search
Software Reliability < Results Home

Click here to complete a short survey. The results of this survey will be used to help us improve the research program and this website.

Click here to view research projects that had new research results added in the last 90 days.
Sensitivity Analysis of Reliability for Structure-Based Software via Simulation
Point of Contact Feng Yang
feng.yang@mail.wvu.edu
Dates May 2007 - December 2007
Problem Computer simulation is an appealing approach for the reliability analysis of structure-based software systems as it can accommodate important complexities present in realistic systems. However, detailed simulation models that reflect the actual execution process of software applications are yet to be developed; and furthermore, the computationally-intensive and stochastic nature of simulation calls for sophisticated statistical methodologies to ensure the efficiency and validity of the analysis. The research proposed here has two complementary threads: developing simulation models to represent structure-based software systems, and performing sensitivity analysis on software reliability based on efficient experimental design and statistical output analysis of simulation experiments.
Objective The objective of this research is to develop a simulation-based methodology for reliability sensitivity analysis of structure-based software systems. Therefore, the criterion for research success lies in the ability of simulation models to provide accurate reliability estimates for software systems, and the efficiency of estimating the impact of various parameters on reliability. The following tasks will be performed in year 2007:
  • Jan.-Feb.: Develop a framework for simulation modeling of structure-based software systems.
  • Mar.-April: Construct the simulation model; Test and verify the model through numerical experiments.
  • June-Dec.: Develop the statistical methodology for the sensitivity analysis via simulation, and perform an efficiency study of the proposed methods.
    • Results SAS_07_Exec_Brief_Sensitivity_Analysis_Simulation_Yang.ppt
    SAS_07_Tech_Pres_Sens_Anal_Rel_SBS_Yang.ppt
    Conference Presentation - Controlled Sequential Bifurcation for Software Reliability Study.pdf
    Keywords Software reliability, sensitivity analysis, discrete-event simulation, experimental design
    Categories Software Reliability


    Curator: Josh Stonestreet
    NASA Official: Lisa Montgomery
    		 NASA Insignia NASA Privacy, Security, Notices