Design And Implementation Of Software Reliability Model Evaluation Decision System

With the rapid development of the Internet industry and computer technology,computer systems and software applications have been widely used in many fields of national defense construction and the national economy,bringing great convenience to people's work and life,and also promoting the rapid development of the country and society.With the increasing demand of software and the exponential growth of its function and complexity,how to ensure the quality of software is the key research content in the field of software reliability engineering.The software reliability model is a measure of confidence in the correctness of software operation,an effective tool for quantifying software reliability,speculating on the future behavior of software,and guiding testing plans to ensure software quality.The software reliability model is based on the abstract assumption of the real test environment.For the convenience of mathematical modeling,the assumption conditions consider the parameters with some emphasis.The selection of a suitable software reliability model for a particular testing environment has become the focus of current research in the field of software reliability.In this paper,the software reliability model is taken as the research object,and the selection and ranking of software reliability model are studied in depth based on the multi-attribute decision making method.This article first analyzes the impact of testing coverage and testing effort on software reliability,and then constructs a model framework considering testing coverage and testing effort under the imperfect debugging mechanism.The analysis and construction process is as follows: 1)Testing coverage is an effective measurement of test efficiency and test adequacy.It analyzes the effect of constructing models under perfect debugging mechanism and imperfect debugging mechanism to evaluate software reliability;2)Testing effort is the measurement of testing resources consumption during the testing process.Also it analyzes the impact of constructing models under perfect debugging mechanism and imperfect debugging mechanism to evaluate software reliability;3)Constructing a model framework that considers testing coverage and testing effort under imperfect debugging mechanism,and combine the new generalized Logistics-type testing coverage function and the Stype testing effort function into the model framework to obtain a new specific model,and verify the adaptability and validity of the model through experiments.Aiming at the application inconsistency of software reliability models,this paper proposes a software reliability model evaluation and decision-making method based on multi-attribute decision-making methods.The specific construction process is as follows: 1)The utility mechanism of software reliability model is analyzed,and the inconsistency problem of software reliability model in application is verified from the perspective of experiment.2)The parameter estimation of Software reliability model is a nonlinear problem,and particle swarm optimization can quickly solve the problem and obtain the global optimal solution.By using the improved particle swarm optimization algorithm combined with the maximum likelihood estimation method to optimize the model parameters,the accuracy of the parameters can be improved,and the prediction performance of the software reliability model can be improved.3)Based on the previous analysis and discussion,a software reliability model evaluation decision-making method based on TOPSIS is proposed for model selection,which comprehensively considers the proximity of the model to the optimal solution and the worst solution,and takes into account the evaluation criterion of weights,which can evaluate the model comprehensively and adequately.Through the experimental comparison with other methods,the adaptability and superiority of the method are verified.Finally,based on the research results of this subject,a software reliability model evaluation and decision-making system is designed and implemented.The system can verify the fitting performance and prediction performance of the new model,and obtain the best-performing model which performs well on the uploaded or sorted failure data set.It also improves the work efficiency of researchers,and provides guidance for subsequent software testing resource management and optimal release.