Development of an Environment for Software Reliability Model Selection

Abstract

An environment was developed for solving the problem of selecting a proper software reliability model for a given set of software failures. The idea behind the environment developed in this dissertation was to liken a collected set of software failure data to a previous one that proved to fit well with a specified software reliability model. Software failures were assumed to have a two-Parameter exponential distribution with unequal type II censoring. A test criterion was derived for testing Ratio Criterion. The exact distribution of the test criterion was derived. An asymptotic approximation was also obtained and was found to be very close to the exact distribution when the number of failures were more than twenty. Software failure data, available from Data and Analysis Center for Software (DACS), were used as the initial group of software failure sets. The environment was then applied, for testing the quality of several software failure sets.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1992
Accession Number
ADA256609

Entities

People

  • Salah A. Elewa

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • C4I
  • Energy and Power Technologies
  • Space

DTIC Thesaurus Topics

  • Air Force
  • Bayesian Networks
  • Computational Science
  • Computer Programming
  • Computer Programs
  • Computers
  • Data Analysis
  • Databases
  • Debugging
  • Information Science
  • Probabilistic Models
  • Probability
  • Probability Distributions
  • Random Variables
  • Reliability
  • Software Development
  • Software Metrics

Fields of Study

  • Computer science
  • Engineering

Readers

  • Regression Analysis.
  • Software Engineering.