Home > Engineering essays > Software Reliability Prediction

Essay: Software Reliability Prediction

Essay details and download:

  • Subject area(s): Engineering essays
  • Reading time: 12 minutes
  • Price: Free download
  • Published: 7 June 2012*
  • File format: Text
  • Words: 3,583 (approx)
  • Number of pages: 15 (approx)

Text preview of this essay:

This page of the essay has 3,583 words. Download the full version above.

Software Reliability Prediction

Zigmund Bluvband (2011) has described two advanced analytical models which are used for obtaining accurate results for software reliability prediction. The First model as described is shown to inhibit some of the specific features of the software testing process and it is based on the well-known S shaped Ohba model. This advanced model can only be applied for the non-rare bug testing. For the rare bug rate prediction another model has been proposed which is based on the introduction of the additional control parameter last suspended time. He has described the advanced parametric models which can be used for the assessment and prediction of software reliability that is to find the bugs present in the software which is based on the statistics of the bugs that are found during the initial stage of testing. The parametric model approach is commonly associated with the reliability issues which are mainly used for dealing with the evaluation of the amount of the bugs found in the code.
Computed parameter from the allowed estimating:
Number of bugs that are remaining in the product
Time required for detecting the remaining bugs.
In his paper he describes different models that are developed for achieving the same objective; these are the Duane Reliability Growth Model, Goel Model, Weibull Model, Classical S-shaped Model, Ohba S-shaped Model, etc. By observing some details of these models and by working on the practical aspects of the software testing process a few Advanced Models were developed and the same have been implemented. The model which has been proposed is observed to be sensitive to the situations typical seen as for the early stages of Software development. This one deals with the essentially non-linear, multimodal goal function and is used to define the optimal value as for the estimation of the unknown control parameter. In his research paper he has described the advanced parametric models for assessment and prediction of software reliability which is based on the statistics of bugs found during the initial stage of testing. The parametric model approach, which is known to be commonly associated with the reliability issues, it deals with the objective of evaluation of the amount of bugs in the code. The proposed models are observed to be sensitive to the situations which are typical for the early stages of Software development. As a result, one deals with the essentially non-linear, multimodal goal function to define the optimal value as the estimation of the unknown control parameter. To support the optimization of such complex models, the Cross-Entropy Global Optimization Method is proposed. Also ome authentic numerical examples have been observed to demonstrate the efficiency of the proposed models.
Akira Hada,(2011) describes in his research paper various models that are used to estimate the reliability for a future profile with increased stress using the current observations to develop a model for future reliability.

For the foreseeable future the equipment itself will not change, and thus if the system is exposed to any increase on the load or stress this may decrease the reliability and increase the maintenance of the system. The model is thus required to determine the impact that these future profiles will have on the system.
This paper was affected by an actual application modelling system and has been observed for the current and the most anticipated future reliability of the equipment used for launch of naval aircraft and recovery. The System reliability models have been presented and are used for demonstration that is used to anticipate the system reliability and its availability required to change and increase the applied loading distributions. In his research paper he has described that these models are intended and best suited for those systems whose components are well exposed to the predictable and quantifiable, but the continuous changing loading patterns. In the described application, the models that will be used to estimate the future reliability of the naval aircraft equipments and the arresting gear when are subjected to various different air wing compositions.
In this each air wing usage profile is potentially different, thus forming a distribution of these usage stresses, and this distribution is observed to be shifting with time, as the weight of aircraft and the missions change. The current ALRE systems and the model which is required to predict the future performance of the systems and to identify the most unreliable components of the systems or problems that are persisting within the existing design; will continue to be used for the foreseeable future. In his research paper he defined that Weibull distribution models that are used in the typical fashion to model the component failure times, but the initial Weibull distribution parameters that have been depicted and observed are the mathematical functions of the current, and the well known applied stress distributions.
Kuei-Chen Chiu (2011) has described that it has been observed that over the last two decades, various software reliability growth models (SRGM) have been proposed, also it has been seen that there has been a gradual but a remarkable and marked shift in the balance between the software reliability and software testing cost in recent years.
Shuanqi Wang(2011) has described that in order to incorporate the effect of test coverage, two software reliability growth models (SRGMs) are proposed in this paper by using and studying the failure data and the test coverage simultaneously. When using the testing time it is found that among these two one is found to be continuous, and the other is found to be discrete when studying it with respect to the number of executed test cases instead of studying it with respect to the testing time like the previous one. Since it has been observed that one of the most important factors of the coverage-based SRGMs is the test coverage function (TCF), the discrete TCF based on Beta function has been discussed first.
Then Shuanqi Wang developed mean value functions (MVF) of the two models thus integrating the two important aspects of testing the system that are the test coverage and imperfect debugging. And in the final stage the proposed TCF and MVFs are evaluated and validated based on the actual software reliability data that is collected from some real software development projects. The observed results clearly demonstrate that both the proposed TCF and SRGMs provide better estimation and fitting for the data sets that are used under comparisons.
She describes that recently an important trend that is used in developing SRGMs is incorporating some other additional information such as test coverage into the SRGMs. This is emerging because that it is seen that only the faults located that are covered in the constructs of the software can only be exposed during the testing process. In this paper, two novel NHPP SRGMs approaches based on the aspect of test coverage have been proposed to describe the relationship between reliability and test coverage,
In this he has discussed a discrete TCF with respect to the number of test cases that are based on the Beta function. Then we have to develop discrete and continuous SRGMs by integrating the aspect of test coverage. The proposed TCF and SRGMs have been evaluated and validated on one published failure data set and a testing project. It is found that for the data sets under comparisons, both the proposed TCF and SRGMs provide a better reliability fitting and estimation power It shows that the proposed approach of software reliability modelling incorporating test coverage is an effective and a successful attempt, which may prove to be very promising for use in some further researches and can be used for developing various applications.
Sultan Aljahdali performed a work,’ Improved Software Reliability Prediction through Fuzzy Logic Modelling’. This paper presents a new approach to assess the software reliability by using the aspect of Fuzzy logic. There is a series of Fuzzy logic modelling which is found to be associated with the different time segments that can be directly used as a piecewise linear model for assessing the reliability and identifying the problems, which can produce some meaningful results in the early stages of the testing process. The model has been applied to three different applications and the fuzzy logic and the normalized root mean of the square of error have been used as an evaluation criterion. Also the results show that the fuzzy model that has been adopted has a good predictive capability.

Sultan Aljahdali has depicted some work on ,’Predicting the Reliability of Software Systems Using Fuzzy Logic’. In this paper, he has explored and worked on the use of fuzzy logic to build a SRGM. The proposed fuzzy model consists of a collection of linear sub-models which have been joined together smoothly using the fuzzy membership functions which are used to represent the fuzzy model. The Results and analysis based data set that has been developed by John Musa of Bell Telephone Laboratories shows the potential advantages of using the fuzzy logic in solving this problem.
Khalaf Khatatneh has provided some work on,’ Software Reliability Modelling Using Soft Computing Technique’. In this paper, a model has been explored that can be used for software reliability prediction. The proposed model that has been proposed in this paper is implemented using the fuzzy logic technique and it has been applied on a collected custom set of test data. The model has been observed to be characterized as a growth reliability model. This model has focused on some particular dataset behaviour that is used for predicting the reliability. The focus has been laid on particular dataset behaviour so that an accurate model can be developed since the recent works have been observed to focus on developing a model which can be more accurate in predicting the results and obtaining outputs.
GUO JUNHONG has performed work on,’ Software Reliability Nonlinear Modelling and Its Fuzzy Evaluation’. This paper presents a nonlinear model of software reliability which is based on the time series and also it gives the corresponding algorithms. The simulation experiments that were performed depicted the accuracy and efficiency of this new model. The newly proposed model is observed to suit the requirements of software engineering better and also the parameters of this model can reflect the changing of the software reliability in a system. The new model is shown to provide a more accurate analysis and to forecast the software reliability issue without making any strict assumptions.
Sultan H. Aljahdali has performed work on,’ Employing four ANNs Paradigms for Software Reliability Prediction: an Analytical Study’. In this paper he has explored the connectionist artificial neural networks models which provide an alternative approach which is used to derive these models by observing the performance analysis of four different connectionist paradigms which are used for modelling the prediction of software reliability. The four paradigms that have been presented are based on the multi-layer perceptron neural network, radial-basis functions, Elman recurrent neural networks and a Takagi- Sugeno fuzzy inference system which are learned using a neural network algorithm (neuro-fuzzy model).
Ajeet Kumar Pandey has performed work on,’ Fault Prediction Model by Fuzzy Profile Development of Reliability Relevant Software Metrics’. In paper a fault prediction model has been presented using the software metrics and fuzzy inference system that are reliability relevant. This new approach has been discussed to develop a fuzzy profile of the software metrics which have been observed to be more relevant for the prediction of software faults. The proposed model with the help of relevant software metric predicts the fault density at the end of each phase of software development.
K. Krishna Mohan has performed work on, ‘Selection of Fuzzy Logic Mechanism for Qualitative Software Reliability Prediction’. The purpose of this work is to demonstrate the mechanism of fuzzy logic for the qualitative prediction of Software Reliability. The Validation of this approach can be easily obtained by comparing the results of this with those results which have been obtained on the realized prototypes at the module level. This work has not been done earlier and this involves studying at the PoC level, qualitative predictions which are used for the metric ‘number of defects’ which can be obtained using a generic Fuzzy Logic based on the modelling.
Jie Yang has performed work on, ‘Managing knowledge for quality assurance: an empirical study’. The main purpose of this paper is to examine the relationship between the two aspects of knowledge management and the quality of a new product which are used to identify the different hidden patterns in which knowledge acquisition and the dissemination of knowledge can affect the quality of a new product. It employs the concept of analysing the Additivity and Variance Stabilization (AVAS). Findings ‘ The quality of new product is related to knowledge management significantly.
Michael R. Lyu performed a work, ‘Optimal Allocation of Test Resources for Software Reliability Growth Modelling in Software Development’. This paper is shown to consider the ‘software component testing resource allocation’ for a system which has single or multiple applications, each system having a pre-described requirement for reliability. The relation between the failure rates of components and the ‘cost to decrease this rate’ has been modelled by using various types of growth curves for reliability. The solutions to the problem in closed-form for the systems with a one single application are developed, and then description has been provided about how to solve the multiple application problem using nonlinear programming techniques. Also the interactions between the system components and the dependencies on the inter-component failures that are included in the modelling formula have been closely examined.
A. Yadav has performed work on,’ Critical Review on Software Reliability Models’. There has been a lot of work done on the estimation of software reliability. Some of the major models that have appeared in the literature are discussed in this paper. The taxonomy on software reliability models has been presented as a major contribution. Reliability models based on the various dimensions have been observed. The models which have been under review are reflecting either an infinite or finite number of failures. All the models based on exponential distribution reflect a finite amount of failures. Also on the other hand the model based on the logarithmic distribution reflects an infinite amount of failures.
Michael R. Lyu has performed work on,’ Optimization of Reliability Allocation and Testing Schedule for Software Systems’. In this paper the author has considered the problem of allocation of software component for a system which has multiple applications. The failure rates of the components which are used to build the applications are related to the testing cost through various types of growth curves of software reliability. The author has achieved solutions in closed-form to the problems where there is only one single application in the system. The analytical solutions that are not readily available where there are multiple applications; however, the numerical solutions can be easily obtained using a non-linear programming tool.
J. O. Omolehin performed a work, ‘Graphics to fuzzy elements in appraisal of an in-house software based on Inter-failure data analysis’. The performance of any software can be approximately measured by using these three parameters; Reliability, Availability and Maintainability. These parameters provide a good amount of information about the robustness of the software which is under the consideration. In this work, software is developed to computerize the results of students of the University of Ilorin and has been designated as in-house software.
P.C. Jha has performed work on, ‘A fuzzy approach for optimal selection of COTS components for modular software system under consensus recovery block scheme incorporating execution time’. The author has proposed a comprehensive approach that can be used to systematically evaluate each component against the selection criteria for functionality, execution time, fault tolerance and quality attributes. The author has developed a model which enables the selection of COTS components by maximizing the reliability and minimizing the absolute deviational execution time which in turn minimizes the overall execution time of the software system
Qiuying Li performed work on,’ A Software Reliability Evaluation Method’. In this paper the method which has been proposed provides information about the software’s reliability at every stage which can be taken as the reference or in accordance to guide the stages of software’s design, analysis and testing and so on. The software reliability evaluation method that has been put forward in this paper, focuses on lots of information which is correlated to the reliability during the whole software life cycle. Also an application has been put forward to demonstrate the feasibility of this method.
N. Raj Kiran has performed work on, ‘Software reliability prediction by soft computing techniques’. In this paper, some models are developed to accurately forecast the software reliability. Various statistical (multiple linear regression and multivariate adaptive regression splines) and intelligent techniques (back propagation trained neural network, dynamic evolving neuro’fuzzy inference system and TreeNet) constitute the ensembles that have been presented.
B. Hailpern has performed work on, ‘Software debugging, testing, and verification’. The author has observed that due to the informal nature of software development as a whole, the practices that are prevalent in the software industry are still immature, even in the areas where the improved technology still exists. In addition to that the, tools that incorporate the more advanced aspects of this technology are not yet ready for large-scale commercial use. Hence there is still reason to hope for some significant improvements in this area over the next several years.
Kevin K.F. Yuen has performed work on, ‘Evaluating Software Quality of Vendors using Fuzzy Analytic Hierarchy Process’. This paper has proposed a fuzzy Analytic Hierarchy Process model for evaluating the software quality of the software provided by the vendors. The criteria of software quality adopt the international norm ISO/IEC9126-1:2001 which consists of six criteria. The fuzzy AHP model applies the modified fuzzy Logarithmic Least Squares Method (LLSM) in this software criteria models. The proposed model can help the developers and testers to evaluate the software applications of the vendors and select the best alternative under any uncertain environment.
ALBEANU G. has performed work on, ‘NTUITIONISTIC FUZZY METHODS IN SOFTWARE RELIABILITY MODELLING’. This paper has described the framework of component-based software which illustrates the usage of intuitionistic fuzzy numbers in imprecise software reliability modelling and computing. It is shown that how intuitionistic triangular and trapezoidal fuzzy numbers can be used for computing the intuitionistic fuzzy reliability of serial, parallel and hybrid software architectures.
Dr.Seetharam .K. has performed work on, ‘Implementation of Multivariate Clustering Methods for Software Development’. In this paper Qualitative/quantitative measurement of software has been discussed using the concept of cluster analysis. In this paper three different cases have been described and worked on. First analysis is done with size as predominant factor, the second analysis is done with effort as predominant factor and the third analysis is done with duration as predominant factor, in the list of eight factors with software reliability performance.
Er. Kailash Aseri has performed work on, ‘A Mathematical Study of Fuzzy Logic Techniques in Software Engineering Measurements’. The estimation models in software engineering are used to predict some important attributes of the future entities such as software development effort, software reliability and programmer productivity. Estimation by fuzzy logic techniques is a very useful technique in field of software effort estimation. In this paper the author has proposed a new approach based on the reasoning by fuzzy logic to estimate effort.
Andrew R. Gray has performed on work, ‘Applications of Fuzzy Logic to Software Metric Models for Development Effort Estimation’. Despite the financial benefits from developing accurate and usable models, there are a number of problems prevailing that have yet not been overcome using the traditional techniques of formal and linear regression models. These include the nonlinearities and interactions that are still inherent in complex real world development processes, the lack of stationary in such processes, over-commitment to precisely specified values, the small quantities of data often available, and the inability to use whatever knowledge is available where exact numerical values are unknown.
Scott Dick has performed work on, ‘Software-Reliability Modeling: The Case for Deterministic Behaviour’. In this paper study has been done on two sets of real-world software reliability data using the techniques of chaotic time-series analysis. In this paper it has been found that both appear to arise from a deterministic process, rather than a stochastic process, and that both show some evidence of chaotic dynamics. In addition, in this paper work has been conducted on a series of k-steps-ahead forecasting experiments in the datasets, pitting a number of well-known stochastic SRMs against radial basis function networks (RBFNs), which are deterministic in nature. The out-of-sample prediction results from the RBFNs showed an improvement of roughly 25% over the best of the stochastic models, for both of Presented datasets. Finally, Author propose a causal model to explain these results, which hypothesizes that faults in a program are distributed over a fractal subset of the program’s input space.
H.Nematzadeh performed a work,’ Evaluating Reliability of System Sequence Diagram Using Fuzzy Petri Net’. Since UML is semi formal, many researches and effort have been performed to transform this language in to formal methods including Petri nets. Thus, the operation of verification and validation of the qualitative parameters could be achieved with more accuracy. Since the majority of the real world information is uncertain, therefore fuzzy UML diagram has been extensively used by system analyzer. This paper attempts to transform system sequence diagram created in fuzzy UML into fuzzy Petri net. Then the reliability is calculated.

...(download the rest of the essay above)

About this essay:

If you use part of this page in your own work, you need to provide a citation, as follows:

Essay Sauce, Software Reliability Prediction. Available from:<https://www.essaysauce.com/engineering-essays/software-reliability-predication/> [Accessed 29-03-24].

These Engineering essays have been submitted to us by students in order to help you with your studies.

* This essay may have been previously published on Essay.uk.com at an earlier date.