Home // ICSEA 2014, The Ninth International Conference on Software Engineering Advances // View article

Low-Variance Software Reliability Estimation Using Statistical Testing

Authors:
Fouad ben Nasr Omri
Safa Omri
Ralf Reussner

Keywords: Software reliability testing, reliability estimation, statistical testing, stratified sampling

Abstract:
Software statistical testing establishes a basis for statistical inference about the expected field quality of software based on an expected operational profile. The standard statistical testing approach draws randomly test cases from the expected operational profile according to the statistical distribution on the expected inputs. Standard Statistical testing is in the most of the cases impractical. The number of required test cases to reach a target reliability with a given confidence is too large. In this paper, we present a test selection approach to minimize the variance of reliability estimator and reduce the overhead of estimating reliability. The presented approach combines the idea of statistical testing with the concepts of stratified sampling. Experiments are conducted to validate the efficiency of our approach.

Pages: 286 to 292

Copyright: Copyright (c) IARIA, 2014

Publication date: October 12, 2014

Published in: conference

ISSN: 2308-4235

ISBN: 978-1-61208-367-4

Location: Nice, France

Dates: from October 12, 2014 to October 16, 2014