n-Steps ahead software reliability prediction using the Kalman filter

Author(s)

Ursini, Edson L. ^[1] ; Martins, Paulo S. ^[1] ; Moraes, Regina L. ^[1] ; Timoteo, Varese S. ^[1]

Affiliation

^[1] Univ Estadual Campinas, UNICAMP, Sch Technol, BR-13484332 Limeira, SP - Brazil

• Document type: Journal article
• Source: Applied Mathematics and Computation; v. 245, p. 116-134, OCT 15 2014.
• Web of Science Citations: 4

Abstract

This paper presents KSL, a new software reliability growth model (SRGM) based on the Kalman filter with a sub filter and the Laplace trend test. We applied the model to the Linux operating system kernel as a case study to predict the absolute and relative (per lines of code) number of faults n-steps ahead. The Laplace trend test is applied to detect when the series no longer follows a homogeneous Poisson process, improving the confidence level. An example is provided with a prediction of 13 months ahead on the number of faults with 8% error. The results (i.e. predictive capability) indicated that the proposed approach outperforms the S-shaped prediction model, Weibull, and Exponentiated Weibull distributions, as well as typical and OS-ELM Neural networks when the series has a short number of observations. (C) 2014 Elsevier Inc. All rights reserved. (AU)

• FAPESP's process: 13/17823-0 - FAULT CLASSIFICATION IN MOBILE CLOUD COMPUTING
• Grantee: Regina Lúcia de Oliveira Moraes
• Support Opportunities: Regular Research Grants
• FAPESP's process: 11/17339-5 - Methodology for Analytical, Simulation and Emulation Models to Dimensioning IP VPN (Internet Protocol -Virtual Private Networks).
• Grantee: Edson Luiz Ursini
• Support Opportunities: Regular Research Grants