Proceedings of the First International Conference on Information Technology and Knowledge Management, 2018
Reliability Allocation in an essential task of the software development process. Increasing compl... more Reliability Allocation in an essential task of the software development process. Increasing complexities in software structure and demand for bug free software has made Reliability Allocation a mandatory task during design and planning phase. So far in the literature several methods and models have been discussed for achieving the reliability target based on user's and developer's point of view. The crucial question that arises is 'How to allocate reliability for a Software system in an uncertain atmosphere where developer's preferences are subjective in nature?". In this paper, we have proposed the software reliability allocation problem incorporating the decision maker's subjective uncertain preferences using Ordered Weighted Averaging (OWA) approach based on Fuzzy Analytical Hierarchical Process (FAHP).Parameter determination using FAHP through architectural hierarchy of the software system helps in interaction of user's assessment with the software engineers and programmers outlook. The OWA technique ensures complete use of available information and also avoids any kind of biasedness in reliability allocation due to overestimation of developer's inclinations. The proposed MEMV-OWA (Maximum Entropy Minimum Variance) operator is a bi-objective mathematical programing problem that maximizes entropy (deployment of information) along with minimization of the variance in weighting vector in an uncertain environment. Reliability allocation procedure for software system using the anticipated process has been discussed in detail. Also precise demonstration of the procedure has been done with an application example.
Software development team intent to provide the best quality products under time and resource con... more Software development team intent to provide the best quality products under time and resource constraints. They single out their product in the market using various tools like advertisements, discounts, free trials, social networking etc. Warranty is one such tool used for product promotion. It assures customers from failure risks after product release. On one hand this helps to build the customer trust and brand image while on the other side it involves cost to developers in terms of cost incurred due to fault corrections during warranty period. Hence management has to maintain a balance so that they don’t have to bear undue financial losses. Therefore they plan release policies to determine the release time and the warranty time period that minimizes the cost. In this paper we have proposed a cost optimization model considering FRF based Two-Dimensional Delayed S-Shaped SRGM to define the failure behaviour under imperfect debugging, market opportunity cost, environmental factor and warranty policy. The model differentiates between the number of failures experienced and faults removed along with the probability of error generation. Once software is released it is affected by various external factors that may affect its failure behaviour. To accommodate the effect of external factor we have introduced an environment factor in the cost model. We have validated the model using the Tandem computers failure dataset. The optimization problem has been solved in MATLAB to determine optimal warranty and release time periods so that the total expected cost for software development is minimized.
Numerous Non-Homogenous Poisson Process based Software Reliability Growth Models have been develo... more Numerous Non-Homogenous Poisson Process based Software Reliability Growth Models have been developed in the past to assess reliability growth of the software system. In a contribution to existing literature, we study the impact of realistic factors encountered during the process of development. This paper proposes a reliability growth model incorporating Fault Reduction Factor, Fault Removal Efficiency and error generation to predict the reliability of the product released in multiple versions. Fault Reduction Factor has been modelled by Delayed S-shaped model, while Fault Removal Efficiency and Error Generation parameters are considered to be constant. Further, we validate the model on a four release real fault dataset of Tandem computers and employ data analysis techniques to plot goodness of fit curves and boxplots to investigate predictive accuracy of proposed model. We have also judged the experimental outcomes considering three special cases. It is witnessed that the proposed model well captures the failure phenomenon and the parameters considered significantly impacts the accuracy.
This paper investigates the attributes related to the software development process (SDP) that aff... more This paper investigates the attributes related to the software development process (SDP) that affect software reliability (SR). In addition, the impact of management review (MR) on SR during testing period is studied. An interactive path model is developed to examine interrelationships between SDP factors, MR and SR. Partial Least Square is used for examining the consistency of factors within the model and to test predictive validity based on the hypothesis developed for relationships among factors. The survey-based research study is conducted to validate the model by collecting data from software professionals working at different job positions. The statistical results reveal that there is a direct positive influence of SDP factors on SR and MR positively moderates the relation between testing and SR. This means that management’s frequent assessment of the testing process, together with better planning and execution of SDP components, improves SR.
Advances in Systems Analysis, Software Engineering, and High Performance Computing, 2020
This chapter presents the application of grey wolf optimizer in software release planning conside... more This chapter presents the application of grey wolf optimizer in software release planning considering warranty based on the proposed mathematical model that measures reliability growth of software systems. Hence, optimal release and warranty time is determined while minimizing the overall software development cost. The software cost model is based on failure phenomenon modelled by incorporating fault removal efficiency, fault reduction factor, and error generation. The model has been validated on the fault dataset of ERP systems. Sensitivity analysis has been carried out to study the discrete changes in the cost parameter due to changes in optimal solution. The work significantly contributes to the literature by fulfilling gaps of reliability growth models, release problems considering warranty, and efficient ways for solving optimization problems. Further, the grey wolf optimizer result has been compared with genetic algorithm and particle swarm optimization techniques.
Handbook of Research on Emerging Trends and Applications of Machine Learning, 2020
Software firms plan all development and management activities strategically to provide the best p... more Software firms plan all development and management activities strategically to provide the best products and solutions to their user. IT professionals are involved in the process of studying the bugs reported and assign severity to make decisions regarding their resolution. To make the task fast and accurate, developers use automatic methods. Herein, the authors have used feature selection-based classification technique to decide about the severity of reported bugs. TF-IDF feature selection method is used to select the informative terms, determining the severity. Based on selected terms the support vector machine and artificial neural network classifiers are used for classification. A number of performance measures have been used to test the performance of classification. The bug reports of Eclipse project for JDT and platform products were collected from Bugzilla. The results show that classifying bugs on the basis of severity can be effectively improved by feature selection-based ...
International Journal of Quality & Reliability Management, 2019
PurposeThe purpose of this paper is to identify and quantify the key components of the overall co... more PurposeThe purpose of this paper is to identify and quantify the key components of the overall cost of software development when warranty coverage is given by a developer. Also, the authors have studied the impact of imperfect debugging on the optimal release time, warranty policy and development cost which signifies that it is important for the developers to control the parameters that cause a sharp increase in cost.Design/methodology/approachAn optimization problem is formulated to minimize software development cost by considering imperfect fault removal process, faults generation at a constant rate and an environmental factor to differentiate the operational phase from the testing phase. Another optimization problem under perfect debugging conditions, i.e. without error generation is constructed for comparison. These optimization models are solved in MATLAB, and their solutions provide insights to the degree of impact of imperfect debugging on the optimal policies with respect to...
International Journal of Mathematics in Operational Research, 2019
In the last few decades, a plenty of software reliability growth models with diverse parameters h... more In the last few decades, a plenty of software reliability growth models with diverse parameters have been formulated and proposed to assess software trustworthiness. Fault reduction factor (FRF) is one such significant factor which has been incorporated in numerous reliability modelling researches in past and results have proved its importance on reliability growth phenomenon. To keep pace with rising demands of the market, software development firms keep upgrading their products in terms of new features and by providing rectification of previously reported defects. In this paper, an SRGM based on NHPP integrating imperfect debugging and time-variable FRF is proposed for multi-release software systems. The model is then tested on the failure datasets of multiple releases of Tandem Computers and Firefox OSS. The experimental results for each release are compared with their corresponding perfect debugging model. Results illustrate that the proposed model fits the failure data well, outperforming perfect debugging model.
Introduction: The present research was conducted at the University of Delhi, India in 2017.Method... more Introduction: The present research was conducted at the University of Delhi, India in 2017.Methods: We develop a software reliability growth model to assess the reliability of software products released in multiple versions under limited availability of resources and time. The Fault Reduction Factor (frf) is considered to be constant in imperfect debugging environments while the rate of fault removal is given by Delayed S-Shaped model.Results: The proposed model has been validated on a real life four-release dataset by carrying out goodness of fit analysis. Laplace trend analysis was also conducted to judge the trend exhibited by data with respect to change in the system’s reliability.Conclusions: A number of comparison criteria have been calculated to evaluate the performance of the proposed model relative to only time-based multi-release Software Reliability Growth Model (srgm).Originality: In general, the number of faults removed is not the same as the number of failures experien...
International Journal of Industrial and Systems Engineering, 2021
Systematic management in the early stage of development helps the software firms to develop a qua... more Systematic management in the early stage of development helps the software firms to develop a qualitative software product. One of the crucial tasks in the software design phase of software development life cycle is to assign reliabilities to modules based on the overall reliability goal. AHP has been extensively used to allocate reliability among modules on the basis of preferences given by the decision-maker, who can be an industry expert or senior level IT personnel. Their opinions are obtained in linguistic terms and later converted to crisp values for comparison at various levels. This approach incorporates vague decisions, incomplete information and uncertain preferences during decision-making. In this paper, we discuss an allocation method where preferences are assigned in terms of Pythagorean fuzzy numbers based on a hierarchy that relates the users, software engineers and programmer expectations for the software system. The methodology and concepts used have been discussed in detail. The proposed approach has been implemented through an example problem.
Proceedings of the First International Conference on Information Technology and Knowledge Management, 2018
Reliability Allocation in an essential task of the software development process. Increasing compl... more Reliability Allocation in an essential task of the software development process. Increasing complexities in software structure and demand for bug free software has made Reliability Allocation a mandatory task during design and planning phase. So far in the literature several methods and models have been discussed for achieving the reliability target based on user's and developer's point of view. The crucial question that arises is 'How to allocate reliability for a Software system in an uncertain atmosphere where developer's preferences are subjective in nature?". In this paper, we have proposed the software reliability allocation problem incorporating the decision maker's subjective uncertain preferences using Ordered Weighted Averaging (OWA) approach based on Fuzzy Analytical Hierarchical Process (FAHP).Parameter determination using FAHP through architectural hierarchy of the software system helps in interaction of user's assessment with the software engineers and programmers outlook. The OWA technique ensures complete use of available information and also avoids any kind of biasedness in reliability allocation due to overestimation of developer's inclinations. The proposed MEMV-OWA (Maximum Entropy Minimum Variance) operator is a bi-objective mathematical programing problem that maximizes entropy (deployment of information) along with minimization of the variance in weighting vector in an uncertain environment. Reliability allocation procedure for software system using the anticipated process has been discussed in detail. Also precise demonstration of the procedure has been done with an application example.
Software development team intent to provide the best quality products under time and resource con... more Software development team intent to provide the best quality products under time and resource constraints. They single out their product in the market using various tools like advertisements, discounts, free trials, social networking etc. Warranty is one such tool used for product promotion. It assures customers from failure risks after product release. On one hand this helps to build the customer trust and brand image while on the other side it involves cost to developers in terms of cost incurred due to fault corrections during warranty period. Hence management has to maintain a balance so that they don’t have to bear undue financial losses. Therefore they plan release policies to determine the release time and the warranty time period that minimizes the cost. In this paper we have proposed a cost optimization model considering FRF based Two-Dimensional Delayed S-Shaped SRGM to define the failure behaviour under imperfect debugging, market opportunity cost, environmental factor and warranty policy. The model differentiates between the number of failures experienced and faults removed along with the probability of error generation. Once software is released it is affected by various external factors that may affect its failure behaviour. To accommodate the effect of external factor we have introduced an environment factor in the cost model. We have validated the model using the Tandem computers failure dataset. The optimization problem has been solved in MATLAB to determine optimal warranty and release time periods so that the total expected cost for software development is minimized.
Numerous Non-Homogenous Poisson Process based Software Reliability Growth Models have been develo... more Numerous Non-Homogenous Poisson Process based Software Reliability Growth Models have been developed in the past to assess reliability growth of the software system. In a contribution to existing literature, we study the impact of realistic factors encountered during the process of development. This paper proposes a reliability growth model incorporating Fault Reduction Factor, Fault Removal Efficiency and error generation to predict the reliability of the product released in multiple versions. Fault Reduction Factor has been modelled by Delayed S-shaped model, while Fault Removal Efficiency and Error Generation parameters are considered to be constant. Further, we validate the model on a four release real fault dataset of Tandem computers and employ data analysis techniques to plot goodness of fit curves and boxplots to investigate predictive accuracy of proposed model. We have also judged the experimental outcomes considering three special cases. It is witnessed that the proposed model well captures the failure phenomenon and the parameters considered significantly impacts the accuracy.
This paper investigates the attributes related to the software development process (SDP) that aff... more This paper investigates the attributes related to the software development process (SDP) that affect software reliability (SR). In addition, the impact of management review (MR) on SR during testing period is studied. An interactive path model is developed to examine interrelationships between SDP factors, MR and SR. Partial Least Square is used for examining the consistency of factors within the model and to test predictive validity based on the hypothesis developed for relationships among factors. The survey-based research study is conducted to validate the model by collecting data from software professionals working at different job positions. The statistical results reveal that there is a direct positive influence of SDP factors on SR and MR positively moderates the relation between testing and SR. This means that management’s frequent assessment of the testing process, together with better planning and execution of SDP components, improves SR.
Advances in Systems Analysis, Software Engineering, and High Performance Computing, 2020
This chapter presents the application of grey wolf optimizer in software release planning conside... more This chapter presents the application of grey wolf optimizer in software release planning considering warranty based on the proposed mathematical model that measures reliability growth of software systems. Hence, optimal release and warranty time is determined while minimizing the overall software development cost. The software cost model is based on failure phenomenon modelled by incorporating fault removal efficiency, fault reduction factor, and error generation. The model has been validated on the fault dataset of ERP systems. Sensitivity analysis has been carried out to study the discrete changes in the cost parameter due to changes in optimal solution. The work significantly contributes to the literature by fulfilling gaps of reliability growth models, release problems considering warranty, and efficient ways for solving optimization problems. Further, the grey wolf optimizer result has been compared with genetic algorithm and particle swarm optimization techniques.
Handbook of Research on Emerging Trends and Applications of Machine Learning, 2020
Software firms plan all development and management activities strategically to provide the best p... more Software firms plan all development and management activities strategically to provide the best products and solutions to their user. IT professionals are involved in the process of studying the bugs reported and assign severity to make decisions regarding their resolution. To make the task fast and accurate, developers use automatic methods. Herein, the authors have used feature selection-based classification technique to decide about the severity of reported bugs. TF-IDF feature selection method is used to select the informative terms, determining the severity. Based on selected terms the support vector machine and artificial neural network classifiers are used for classification. A number of performance measures have been used to test the performance of classification. The bug reports of Eclipse project for JDT and platform products were collected from Bugzilla. The results show that classifying bugs on the basis of severity can be effectively improved by feature selection-based ...
International Journal of Quality & Reliability Management, 2019
PurposeThe purpose of this paper is to identify and quantify the key components of the overall co... more PurposeThe purpose of this paper is to identify and quantify the key components of the overall cost of software development when warranty coverage is given by a developer. Also, the authors have studied the impact of imperfect debugging on the optimal release time, warranty policy and development cost which signifies that it is important for the developers to control the parameters that cause a sharp increase in cost.Design/methodology/approachAn optimization problem is formulated to minimize software development cost by considering imperfect fault removal process, faults generation at a constant rate and an environmental factor to differentiate the operational phase from the testing phase. Another optimization problem under perfect debugging conditions, i.e. without error generation is constructed for comparison. These optimization models are solved in MATLAB, and their solutions provide insights to the degree of impact of imperfect debugging on the optimal policies with respect to...
International Journal of Mathematics in Operational Research, 2019
In the last few decades, a plenty of software reliability growth models with diverse parameters h... more In the last few decades, a plenty of software reliability growth models with diverse parameters have been formulated and proposed to assess software trustworthiness. Fault reduction factor (FRF) is one such significant factor which has been incorporated in numerous reliability modelling researches in past and results have proved its importance on reliability growth phenomenon. To keep pace with rising demands of the market, software development firms keep upgrading their products in terms of new features and by providing rectification of previously reported defects. In this paper, an SRGM based on NHPP integrating imperfect debugging and time-variable FRF is proposed for multi-release software systems. The model is then tested on the failure datasets of multiple releases of Tandem Computers and Firefox OSS. The experimental results for each release are compared with their corresponding perfect debugging model. Results illustrate that the proposed model fits the failure data well, outperforming perfect debugging model.
Introduction: The present research was conducted at the University of Delhi, India in 2017.Method... more Introduction: The present research was conducted at the University of Delhi, India in 2017.Methods: We develop a software reliability growth model to assess the reliability of software products released in multiple versions under limited availability of resources and time. The Fault Reduction Factor (frf) is considered to be constant in imperfect debugging environments while the rate of fault removal is given by Delayed S-Shaped model.Results: The proposed model has been validated on a real life four-release dataset by carrying out goodness of fit analysis. Laplace trend analysis was also conducted to judge the trend exhibited by data with respect to change in the system’s reliability.Conclusions: A number of comparison criteria have been calculated to evaluate the performance of the proposed model relative to only time-based multi-release Software Reliability Growth Model (srgm).Originality: In general, the number of faults removed is not the same as the number of failures experien...
International Journal of Industrial and Systems Engineering, 2021
Systematic management in the early stage of development helps the software firms to develop a qua... more Systematic management in the early stage of development helps the software firms to develop a qualitative software product. One of the crucial tasks in the software design phase of software development life cycle is to assign reliabilities to modules based on the overall reliability goal. AHP has been extensively used to allocate reliability among modules on the basis of preferences given by the decision-maker, who can be an industry expert or senior level IT personnel. Their opinions are obtained in linguistic terms and later converted to crisp values for comparison at various levels. This approach incorporates vague decisions, incomplete information and uncertain preferences during decision-making. In this paper, we discuss an allocation method where preferences are assigned in terms of Pythagorean fuzzy numbers based on a hierarchy that relates the users, software engineers and programmer expectations for the software system. The methodology and concepts used have been discussed in detail. The proposed approach has been implemented through an example problem.
Uploads
Papers by VIBHA VERMA