Software Reliability: Modelling Time-to-Error and Time-to-Fix.

Abstract

Various competing software reliability models were tested using time-to-error data extracted from a large data base provided by Rome Air Development Center, (RADC), Griffis AFB, N.Y. Only those models for which appropriate information was available from the RADC data base were used. Each model that was tested had its parameters estimated and discrimination between the models was performed by the Likelihood Ratio Test. Code for the computer program used in parameter estimation and Likelihood Ratio Testing is included. The same large data base was also used in determination of the probability density functions governing the random variable time-to-fix. Time-to-fix data was extracted from the data base and was sorted by error categories. Eleven of the more common probability density functions, along with the Kolmogorov-Smirnov Test and the Likelihood Ratio Test, were used to isolate the distribution that best modelled software time-to-fix for each particular data set.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1978
Accession Number
ADA055227

Entities

People

  • Steve G. Castle

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • C4I
  • Materials and Manufacturing Processes
  • Space

DTIC Thesaurus Topics

  • Air Force
  • Command And Control
  • Computations
  • Computer Programming
  • Computer Programs
  • Computers
  • Data Sets
  • Databases
  • Intervals
  • New York
  • Operating Systems
  • Probability
  • Probability Density Functions
  • Random Variables
  • Reliability
  • Software Development
  • Time Intervals

Fields of Study

  • Engineering

Readers

  • Aerospace Research.
  • Computational Modeling and Simulation
  • Regression Analysis.