Interval Estimations of Optimal Software Release Time Based on MCMC Method  
Author Shinji Inoue

 
Co-Author(s) Shigeru Yamada

 
Abstract As one of the intersting topics in software reliabilty assessment by using a software reliabilty model, there exists a problem on when to ship a software to the users from the economic and quality aspects. This paper discusses an interval estimation method for optimal software release time by applying Markov chain Monte Carlo method to a discretized nonhomogeneous Poisson process (NHPP) model for developing a shipping plan flexibly. Our approach is expected to conduct economic software management in a testing phase. Numerical examples of our approach is shown by using actual data collected from acutal software testing phase.

 
Keywords Software reliability model, Optimal release problem, Interval estimation, Markov chain Monte Carlo method, Discretized NHPP model
   
    Article #:  24121
 
Proceedings ISSAT International Conference on Reliability and Quality in Design 2018
August 2-4, 2018 - Toronto, Ontario, Canada