Software reliability prediction using NHPP and Least median of squares (LMS) for parameters estimation

Authors

  • Mayuri H. Molawade, Dr. Shashank D. Joshi, Dr. Rohini Jadhav

DOI:

https://doi.org/10.17762/msea.v71i3.460

Abstract

Based on Cauchy distributions, this research provides an Innovative Assessment Model for Software Reliability Prediction. In this study, the finite-fault NHPP reliability model is used to analyse the performance of software reliability using Cauchy distributions. The Least Median of Squares (LMS) approach was used to estimate the parameters, and Newton's method was used to solve the nonlinear equations.The Cauchy distributions model therefore performed well. Because the mean square error (MSE) and failure incidence rate both decrease over time, it is important to look at the intensity function. When the software reliability was examined after setting the mission time down the road, the Cauchy distributions model used to have a larger durability trend than another models, which indicates a decline in reliability with mission time. As a result, the Cauchy distributions model outperforms the other exiting models and thus, the improved model used to software developers in order to improve programmereliability.

 

Downloads

Published

2022-08-19

How to Cite

Mayuri H. Molawade, Dr. Shashank D. Joshi, Dr. Rohini Jadhav. (2022). Software reliability prediction using NHPP and Least median of squares (LMS) for parameters estimation. Mathematical Statistician and Engineering Applications, 71(3), 1194–1206. https://doi.org/10.17762/msea.v71i3.460

Issue

Section

Articles