International Journal of Research Studies in Computer Science and Engineering (IJRSCSE)
Volume 2, Issue 3, March 2015, PP 21-24
ISSN 2349-4840 (Print) & ISSN 2349-4859 (Online)
www.arcjournals.org
©ARC Page 21
Burr Type III Software Reliability with SPC-An Order
Statistics Approach
K. Sobhana
Department of Computer Science
Krishna University
Machilipatnam, Andhra Pradesh,India
msobhana@yahoo.com
Dr. R. Satya Prasad
Dept. of Computer Science & Engg
Acharya Nagarjuna University
Guntur, Andhra Pradesh (India)
prof_rsp@gmail.com
Abstract: In the past few decades research on
software reliability has been conducted and several
software reliability growth models have been
developed for estimating software reliability. Order
Statistics is an approach for estimating software
reliability for time domain data based on NHPP with
a distribution model. This paper presents the Burr
Type III model as a software reliability growth model
and derives the expressions for an efficient reliability
function using order statistics. Statistical Process
Control (SPC) can be used to monitor the software
reliability process and thereby improve the software
quality. Control Charts are one of the powerful SPC
tools to analyze the failure frequency. It is proposed
that the SPC can be applied to monitor the software
failure process of Burr Type III based NHPP.
Keywords: Software reliability; Burr type III; Order
Statistics; Statistical Process Control;NHPP
1. INTRODUCTION
Software reliability is one of the most important
characteristics of software quality. Reliable software
systems can be produced and maintained by
employing quality measurement and management
technologies during the software life cycle. Software
Reliability is the probability of failure free operation
of software in a specified environment during
specified time [1].
The monitoring of Software reliability process is a far
from simple activity. In recent years, several authors
have recommended the use of SPC for software
process monitoring. A few others have highlighted
the potential pitfalls in its use[2].
The main thrust of the paper is to formalize and
present an array of guidelines in a disciplined process
with a view to helping the practitioner in putting SPC
to correct use during software process monitoring.
Over the years, SPC has come to be widely used
among others, in manufacturing industries for the
purpose of controlling and improving processes. Our
effort is to apply SPC techniques in the software
development process so as to improve software
reliability and quality [3]. It is reported that SPC can
be successfully applied to several processes for
software development, including software reliability
process. SPC is traditionally so well adopted in
manufacturing industry. In general software
development activities are more process centric than
product centric which makes it difficult to apply SPC
in a straight forward manner.
The utilization of SPC for software reliability has
been the subject of study of several researchers. A
few of these studies are based on reliability process
improvement models. They turn the search light on
SPC as a means of accomplishing high process
maturities. Some of the studies furnish guidelines in
the use of SPC by modifying general SPC principles
to suit the special requirements of software
development [3] (Burr and Owen[4]; Flora and
Carleton[5]). It is especially noteworthy that Burr and
Owen provide seminal guidelines by delineating the
techniques currently in vogue for managing and
controlling the reliability of software. Significantly,
in doing so, their focus is on control charts as
efficient and appropriate SPC tools.
It is accepted on all hands that Statistical process
control acts as a powerful tool for bringing about
improvement of quality as well as productivity of any
manufacturing procedure and is particularly relevant
to software development also. Viewed in this light,
SPC is a method of process management through
application of statistical analysis, which involves and
includes the defining, measuring, controlling, and
improving of the processes[6].
2.1. Model Development
A. NHPP Model
Software reliability probabilistic models can be
classified as Markovian models and fault counting
models. In Markovian model a Markov process
represents the failure process. In fault counting model
the failure process is described by stochastic process
like Homogeneous Poisson Process (HPP), Non
Homogeneous Poisson Process (NHPP) and