Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Post General Availability Software Reliability Estimation Method

IP.com Disclosure Number: IPCOM000100502D
Original Publication Date: 1990-Apr-01
Included in the Prior Art Database: 2005-Mar-15
Document File: 3 page(s) / 75K

Publishing Venue

IBM

Related People

Iwasa, H: AUTHOR

Abstract

Disclosed is a method for Software Reliability Estimation based on Total Software Reliability Growth Model (TSRGM). This method can estimate a software error growth curve, before it is put on the market, when the software product is being developed, as shown in Fig. 1.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 57% of the total text.

Post General Availability Software Reliability Estimation Method

       Disclosed is a method for Software Reliability Estimation
based on Total Software Reliability Growth Model (TSRGM). This method
can estimate a software error growth curve, before it is put on the
market, when the software product is being developed, as shown in
Fig. 1.

      TSRGM is composed of Dual Software Reliability Growth Model
(DSRGM) (*) which is used to measure software reliability in the test
environment and Post GA Software Reliability Growth Model (PGASRGM)
which is used to estimate software reliability on the market.  They
are on a Non-Homogeneous Poisson Process (NHPP) model.  The program
estimates software reliability by using the test data which includes
the run test time and number of errors, test acceleration factor,
test coverage rate and the number of forecasted shipments.

      Fig. 2 shows the error growth curve on the test by DSRGM, and
Fig.  3 shows the error growth curve on the market by PGASRGM.

      The mean value function of DSRGM is as follows:

                            (Image Omitted)

      I (t) = N (1-exp (-bt)/(1+C exp (-bT)).
      C = (1-Q)/Q
      T =  t-d, t-d>O
           O  , t-d<O
where N is the number of errors in a tested program, b is the failure
detected rate, t is the test execution time, C if the inflection
parameter (DSRGM), Q is the inflection rate (DSRGM) which indicates
the r...