A Quantifiable Methodology for Software Testing: Using Path Analysis.

Abstract

This report is a comprehensive presentation of a quantatitive methodology for software testing which measures test effectiveness at several different levels of program coverage and establishes confidence levels in the correctness of the program at these levels. Based on the resulting numerical specifications for testing a computer program, quantatitive acceptance criteria are developed. These metrics are sensitived to cost and software criticality factors. The methodology, based on path analysis, is a natural extension of software engineering techniques to quality assurance for well-structured programs. It has been applied successfully, but several practical problems still remain. Application of this mlethodology to a software development program will provide control and visibility into the structure of the program and may result in improved reliability and documentation. Especially for the Air Force, when it acts only as a monitor, external to the software development process, the methodology provides a framework for proper planning and optimal allocation of test resources by quantifying the effectiveness of a test program and pre-determining the amount of testing required for achieving test objectives. With the proof of the fundamental theorem of program testing in 1975, which establishes testing as the equivalent of a proof of correctness for programs which satisfy some structural constraints, systematic testing has become possibly the only effective means to assure quality of a program of non-trivial complexity.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1981
Accession Number
ADA110196

Entities

People

  • Shashi Phoha

Organizations

  • MITRE Corporation

Tags

Communities of Interest

  • Energy and Power Technologies
  • Materials and Manufacturing Processes
  • Space
  • Weapons Technologies

DTIC Thesaurus Topics

  • Air Force
  • Computer Programming
  • Computer Programs
  • Computer Security Software
  • Computers
  • Contracts
  • Control Systems
  • Databases
  • Debugging
  • Operating Systems
  • Reliability
  • Robotics
  • Software Development
  • Software Testing
  • Structured Programming
  • Test And Evaluation
  • Test Methods

Fields of Study

  • Computer science
  • Engineering

Readers

  • Aerospace Test and Evaluation
  • Applied Combinatorial Optimization and Logic Circuit Design.
  • Computational Modeling and Simulation