Software Reliability Prediction Using Ensemble Learning on Selected Features in Imbalanced and Balanced Datasets: A Review

Redundancy,correlation,feature irrelevance,and missing samples are just a few problems that make it difficult to analyze software defect data.Additionally,it might be challenging to maintain an even distribution of data relating to both defective and non-defective software.The latter software class’s data are predominately present in the dataset in the majority of experimental situations.The objective of this review study is to demonstrate the effectiveness of combining ensemble learning and feature selection in improving the performance of defect classification.Besides the successful feature selection approach,a novel variant of the ensemble learning technique is analyzed to address the challenges of feature redundancy and data imbalance,providing robustness in the classification process.To overcome these problems and lessen their impact on the fault classification performance,authors carefully integrate effective feature selection with ensemble learning models.Forward selection demonstrates that a significant area under the receiver operating curve(ROC)can be attributed to only a small subset of features.The Greedy forward selection(GFS)technique outperformed Pearson’s correlation method when evaluating feature selection techniques on the datasets.Ensemble learners,such as random forests(RF)and the proposed average probability ensemble(APE),demonstrate greater resistance to the impact of weak features when compared to weighted support vector machines(W-SVMs)and extreme learning machines(ELM).Furthermore,in the case of the NASA and Java datasets,the enhanced average probability ensemble model,which incorporates the Greedy forward selection technique with the average probability ensemble model,achieved remarkably high accuracy for the area under the ROC.It approached a value of 1.0,indicating exceptional performance.This review emphasizes the importance of meticulously selecting attributes in a software dataset to accurately classify damaged components.In addition,the suggested ensemble learning model successfully addressed the aforementioned problems with software data and produced outstanding classification performance.

Probabilistic back-analysis of rainfall-induced landslides for slope reliability prediction with multi-source information

Probabilistic back-analysis is an important means to infer the statistics of uncertain soil parameters,making the slope reliability assessment closer to the engineering reality.However,multi-source information(including test data,monitored data,field observation and slope survival records)is rarely used in current probabilistic back-analysis.Conducting the probabilistic back-analysis of spatially varying soil parameters and slope reliability prediction under rainfalls by integrating multi-source information is a challenging task since thousands of random variables and high-dimensional likelihood function are usually involved.In this paper,a framework by integrating a modified Bayesian Updating with Subset simulation(mBUS)method with adaptive Conditional Sampling(aCS)algorithm is established for the probabilistic back-analysis of spatially varying soil parameters and slope reliability prediction.Within this framework,the high-dimensional probabilistic back-analysis problem can be easily tackled,and the multi-source information(e.g.monitored pressure heads and slope survival records)can be fully used in the back-analysis.A real Taoyuan landslide case in Taiwan,China is investigated to illustrate the effectiveness and performance of the established framework.The findings show that the posterior knowledge of soil parameters obtained from the established framework is in good agreement with the field observations.Furthermore,the updated knowledge of soil parameters can be utilized to reliably predict the occurrence probability of a landslide caused by the heavy rainfall event on September 12,2004 or forecast the potential landslides under future rainfalls in the Fuhsing District of Taoyuan City,Taiwan,China.

Reliability Prediction of Wrought Carbon Steel Castings under Fatigue Loading Using Coupled Mold Optimization and Finite Element Simulation

The fatigue life and reliability of wrought carbon steel castings produced with an optimized mold design are predicted using a finite element method integrated with reliability calculations.The optimization of the mold is carried out using MAGMASoft mainly based on porosity reduction as a response.After validating the initial mold design with experimental data,a spring flap,a common component of an automotive suspension system is designed and optimized followed by fatigue life prediction based on simulation using Fe-safe.By taking into consideration the variation in both stress and strength,the stress-strength model is used to predict the reliability of the component under fatigue loading.Under typical loading conditions of 70 kN,the analysis showed that 95%of the steel spring flaps achieve infinite life.However,under maximum loading conditions of 90 kN,reliability declined significantly,with only 65%of the spring flaps expected to withstand the stress without failure.The study also identified a safe load-induced stress of 95 MPa on the spring flap.The findings suggest that transitioning from forged to cast spring flaps is a promising option,particularly if further improvements in casting design reduce porosity to negligible levels,potentially achieving 100%reliability under typical loading conditions.This integrated approach of mold optimization coupled with reliability estimation under realistic service loading conditions offers significant potential for the casting industry to produce robust,cost-effective products.

Evaluating the Impact of Prediction Techniques: Software Reliability Perspective

Maintaining software reliability is the key idea for conducting quality research.This can be done by having less complex applications.While developers and other experts have made signicant efforts in this context,the level of reliability is not the same as it should be.Therefore,further research into the most detailed mechanisms for evaluating and increasing software reliability is essential.A signicant aspect of growing the degree of reliable applications is the quantitative assessment of reliability.There are multiple statistical as well as soft computing methods available in literature for predicting reliability of software.However,none of these mechanisms are useful for all kinds of failure datasets and applications.Hence nding the most optimal model for reliability prediction is an important concern.This paper suggests a novel method to substantially pick the best model of reliability prediction.This method is the combination of analytic hierarchy method(AHP),hesitant fuzzy(HF)sets and technique for order of preference by similarity to ideal solution(TOPSIS).In addition,using the different iterations of the process,procedural sensitivity was also performed to validate the ndings.The ndings of the software reliability prediction models prioritization will help the developers to estimate reliability prediction based on the software type.

Corrosion Fatigue Life Prediction of Aircraft Structure Based on Fuzzy Reliability Approach

Material performance of LY12CZ aluminum is greatly degraded because of corrosion and corrosion fatigue, which severely affect the integrity and safety of aircraft structure, especially those of lbe navy aircraft structure. The corrosion and corrosion fatigue failure process of aircraft structure are directly concerned with many factors, such as load, material characteristics, corrosive environment and so on. The damage mechanism is very complicated, and there are both randomness and fuzziness in the failure process. With consideration of the limitation of those conventional probabilistic approaches for prediction of corrosion fatigue life of aircraft structure at present, and based on the operational load spectrum obtained through investigating service status of the aircraft in naval aviation force, a fuzzy reliability approach is proposed, which is more reasonable and closer to the fact. The effects of the pit aspect ratio, the crack aspect ratio and all fuzzy factors on corrosion fatigue life of aircraft structure are discussed. The results demonstrate that the approach can be applied to predict the corrosion fatigue life of aircraft structure.

Reliability estimation and remaining useful lifetime prediction for bearing based on proportional hazard model

As the central component of rotating machine,the performance reliability assessment and remaining useful lifetime prediction of bearing are of crucial importance in condition-based maintenance to reduce the maintenance cost and improve the reliability.A prognostic algorithm to assess the reliability and forecast the remaining useful lifetime(RUL) of bearings was proposed,consisting of three phases.Online vibration and temperature signals of bearings in normal state were measured during the manufacturing process and the most useful time-dependent features of vibration signals were extracted based on correlation analysis(feature selection step).Time series analysis based on neural network,as an identification model,was used to predict the features of bearing vibration signals at any horizons(feature prediction step).Furthermore,according to the features,degradation factor was defined.The proportional hazard model was generated to estimate the survival function and forecast the RUL of the bearing(RUL prediction step).The positive results show that the plausibility and effectiveness of the proposed approach can facilitate bearing reliability estimation and RUL prediction.

An Approach to Online Reliability Evaluation and Prediction of Mechanical Transmission Components

New development trends in electronic operating data logging systems enable classification, recording and storage of load spectrums of mechanical transmission components during usage. Based on this fact, the application of online reliability evaluation and reliability prediction procedures are presented. Different methods are considered to calculate reliability, depending on actual load spectrum and a Wohler curve. The prediction of a reliability trend is analyzed by the application of time series models. For this purpose, exponential smoothing model, regression model, and the ARIMA model are considered to evaluate data and predict an decreasing reliability trends during usage.

IEC 62059 Standards' Application in Reliability Prediction and Verification of Smart Meters

This article introduces the current situation of the smart then describes the relationship of meter reliability characteristics meter＇s reliability and the failure mechanisms at first, and combined with its Bathtub Curve. It also introduces both the feasible failure tree model for meter lifecycle prediction based on actual experiences and meter reliability prediction methodology by SN 29500 norms based on this model. This article also brings forward that it is necessary that the ＂Learning Factor＂ shall be adopted in meter reliability prediction for new materials, new process, and customized parts by referring to GJB/Z299C. Thereafter, this article also tries to apply IEC 62059 and JB/T 50070 to introduce the feasible method for the lifecycle prediction result verification by accelerated lifecycle test. Furthermore, the article also explores ways to increase the firmware reliability in smart meter.

Markov Chain Modelling of Reliability Analysis and Prediction under Mixed Mode Loading

The reliability assessment for an automobile crankshaft provides an important understanding in dealing with the design life of the component in order to eliminate or reduce the likelihood of failure and safety risks.The failures of the crankshafts are considered as a catastrophic failure that leads towards a severe failure of the engine block and its other connecting subcomponents.The reliability of an automotive crankshaft under mixed mode loading using the Markov Chain Model is studied.The Markov Chain is modelled by using a two-state condition to represent the bending and torsion loads that would occur on the crankshaft.The automotive crankshaft represents a good case study of a component under mixed mode loading due to the rotating bending and torsion stresses.An estimation of the Weibull shape parameter is used to obtain the probability density function,cumulative distribution function,hazard and reliability rate functions,the bathtub curve and the mean time to failure.The various properties of the shape parameter is used to model the failure characteristic through the bathtub curve is shown.Likewise,an understanding of the patterns posed by the hazard rate onto the component can be used to improve the design and increase the life cycle based on the reliability and dependability of the component.The proposed reliability assessment provides an accurate,efficient,fast and cost effective reliability analysis in contrast to costly and lengthy experimental techniques.

Software Reliability Early Prediction in Architectural Design Phase: Overview and Limitations

In recent times, computer based systems are frequently used for protection and control in the various industries viz Nuclear, Electrical, Mechanical, Civil, Electronics, Medical, etc. From the operating experience of those computer based systems, it has been found that the failure of which can lead to the severe damage to equipments or environmental harm. The culprit of this accident is nobody other than our software, whose reliability has not been ensured in those conditions. Also for real time system, throughput of the system and average response time are very important constructs/ metrics of reliability. Moreover neither of the software reliability model is available which can be fitted generically for all kinds of software. So, we can ensure reliability at the early stage i.e. during design phase by architecturing the software in a better way. The objective of this paper is to provide an overview of the state-of-the-art research in the area of architecture-based software reliability analysis. We then describe the shortcomings and the limiting assumptions underlying the prevalent research. We also propose various approaches which have the potential to address the existing

Prediction of Web Services Reliability Based on Decision Tree Classification Method

With the development of the service-oriented computing(SOC),web service has an important and popular solution for the design of the application system to various enterprises.Nowadays,the numerous web services are provided by the service providers on the network,it becomes difficult for users to select the best reliable one from a large number of services with the same function.So it is necessary to design feasible selection strategies to provide users with the reliable services.Most existing methods attempt to select services according to accurate predictions for the quality of service(QoS)values.However,because the network and user needs are dynamic,it is almost impossible to accurately predict the QoS values.Furthermore,accurate prediction is generally time-consuming.This paper proposes a service decision tree based post-pruning prediction approach.This paper first defines the five reliability levels for measuring the reliability of services.By analyzing the quality data of service from the network,the proposed method can generate the training set and convert them into the service decision tree model.Using the generated model and the given predicted services,the proposed method classifies the service to the corresponding reliability level after discretizing the continuous attribute of service.Moreover,this paper applies the post-pruning strategy to optimize the generated model for avoiding the over-fitting.Experimental results show that the proposed method is effective in predicting the service reliability.

Reliability of Plasma Von Willebrand Factor Antigen in Prediction of Esophageal Varices in Patients with Liver Cirrhosis

Background: Bleeding esophageal varices (OVs) due to portal hypertension are one of the major complications with high mortality in liver cirrhosis. So, early detection and management are mandatory. Aim: To evaluate the role of Von Willebrand factor (VWF) in predicting the presence of OVs. Patients and Methods: 62 patients with liver cirrhosis representing different Child-Pugh classes were included. The diagnosis of liver cirrhosis was based on the combination of clinical, laboratory and US examinations. All included patients underwent the following investigations: complete blood count, liver function tests (ALT, AST, serum bilirubin, albumin and total protein, prothrombin time (PT) and concentration (PC), INR and serum alkaline phosphatase), serum creatinine, Von Willebrand factor antigen (VWF-Ag) measurement and abdominal US. Upper endoscopic evaluation was done to detect presence or absence of varices (esophageal or gastric) and/or PHG. Results: 38 males and 24 females with their mean age (46 ± 12 years old) were included. Plasma Von Willebrand factor-Ag level was significantly higher in patients with OVs than those without varices (P value = 0.000). Also, its level was significantly higher in patients with higher grade of OVs, G3 than those with G1 or G2 (P value = 0.000). Patients with large OVs including those with G2 and G3 showed significantly higher values of VWF than those with small OVs (NO and G1) (P value = 0.000). VWF was independent predictor for detecting the presence of OVs with good sensitivity (90), specificity (77.3) and accuracy (85.5) at a cutoff value of 1.74 U/ml. Also it was an independent predictor for detecting the presence of large OVs with good sensitivity (91.2), specificity (85.7) and accuracy (88.7) at a cutoff value of 2.16 U/ml. Conclusion: VWF-Ag could be used as a non invasive laboratory independent predictor for the detection of OVs.

Data Fusion about Serviceability Reliability Prediction for the Long-Span Bridge Girder Based on MBDLM and Gaussian Copula Technique

This article presented a new data fusion approach for reasonably predicting dynamic serviceability reliability of the long-span bridge girder.Firstly,multivariate Bayesian dynamic linear model(MBDLM)considering dynamic correlation among the multiple variables is provided to predict dynamic extreme deflections;secondly,with the proposed MBDLM,the dynamic correlation coefficients between any two performance functions can be predicted;finally,based on MBDLM and Gaussian copula technique,a new data fusion method is given to predict the serviceability reliability of the long-span bridge girder,and the monitoring extreme deflection data from an actual bridge is provided to illustrated the feasibility and application of the proposed method.

Model of the Reliability Prediction of Masonry Walls in Buildings

The suitable repair forecasting is needed for proper maintenance of the buildings. The appropriate maintenance planning should be based on the prognostic analysis of the repair needs. However, in Poland, maintenance planning is currently not seen as a long-term system. Repairs are understood as extemporary work and are carried out exclusively on the basis of intermittent inspections and controls. One of the numerous factors determining maintenance planning is exploitation reliability conditioned by durability. This article presents a proposal to determine the prediction of operational reliability of the building constructed using traditional technology. The method of behaving and changing the reliability of the building throughout its use will be useful in planning renovations. The presented analysis includes apartment buildings erected in a traditional technology and regards them as technical objects. For such approached buildings, it is proposed to apply rules applied for mechanical and electrical objects. The probability of the exploitation of a building without any breakdowns in a given period of time is defined as exploitation reliability.

Mechanical Dispatch Reliability Prediction for Civil Aircraft Considering Operational Parameters

To effectively predict the mechanical dispatch reliability(MDR),the artificial neural networks method combined with aircraft operation health status parameters is proposed,which introduces the real civil aircraft operation data for verification,to improve the modeling precision and computing efficiency.Grey relational analysis can identify the degree of correlation between aircraft system health status(such as the unscheduled maintenance event,unit report event,and services number)and dispatch release and screen out themost closely related systems to determine the set of input parameters required for the prediction model.The artificial neural network using radial basis function(RBF)as a kernel function,has the best applicability in the prediction of multidimensional,small sample problems.Health status parameters of related systems are used as the input to predict the changing trend ofMDR,under the artificial neural network modeling framework.The case study collects real operation data for a certain civil aircraft over the past five years to validate the performance of the model which meets the requirements of the application.The results show that the prediction quadratic error Ep of the model reaches 6.9×10−8.That is to say,in the existing operating environment,the prediction of the number of delay&cancel events per month can be less than once.The accuracy of RBF ANN,BP ANN and GA-BP ANN are compared further,and the results show that RBF ANN has better adaptability to such multidimensional small sample problems.The efforts of this paper provide a highly efficientmethod for theMDR prediction through aircraft system health state parameters,which is a promising model to enhance the prediction and controllability of the dispatch release,providing support for the construction of the civil aircraft operation system.

AN ESTIMATION OF FUZZY RELIABILITY OF DISTINCT ELEMENT METHOD IN PREDICTION OF SURFACE SUBSIDENCE DUE TO COAL MINING

The quantitative evaluation of errors involved in a particular numerical modelling is of prime importance for the effectiveness and reliability of the method. Errors in Distinct Element Modelling are generated mainly through three resources as simplification of physical model, determination of parameters and boundary conditions. A measure of errors which represent the degree of numerical solution 'close to true value' is proposed through fuzzy probability in this paper. The main objective of this paper is to estimate the reliability of Distinct Element Method in rock engineering practice by varying the parameters and boundary conditions. The accumulation laws of standard errors induced by improper determination of parameters and boundary conditions are discussed in delails. Furthermore, numerical experiments are given to illustrate the estimation of fuzzy reliability. Example shows that fuzzy reliability falls between 75%-98% when the relative standard errors of input data is under 10 %.

Assessment and Prediction of Software Reliability in Mobile Applications

Software reliability is an important quality attribute, and software reliability models are frequently used to measure and predict software maturity. The nature of mobile environments differs from that of PC and server environments due to many factors, such as the network, energy, battery, and compatibility. Evaluating and predicting mobile application reliability are real challenges because of the diversity of the mobile environments in which the applications are used, and the lack of publicly available defect data. In addition, bug reports are optionally submitted by end-users. In this paper, we propose assessing and predicting the reliability of a mobile application using known software reliability growth models (SRGMs). Four software reliability models are used to evaluate the reliability of an open-source mobile application through analyzing bug reports. Our experiment proves it is possible to use SRGMs with defect data acquired from bug reports to assess and predict the software reliability in mobile applications. The results of our work enable software developers and testers to assess and predict the reliability of mobile software applications.

Prediction Reliability of the Building Components as a Model Based on the Distribution of Weibull

All technical objects are at risk of damages during the consecutive years of their usage. Reliability of an object is an essential issue during its usage. The main problem is the strive to eliminate damage formation. Predicting the reliability of an object should allow qualitative and quantitative analysis of the possibility of occurrence of unfavorable events. The adaptation of mathematical models describing the degradation processes in mechanical and electronic devices creates opportunities to develop diagnostic standards for buildings erected in traditional technology. The article presents the methodology of prediction of reliability of a building, and the values of performance features are defined by the parameters of the Weibull distribution function.

High Cycle Fatigue Life Prediction and Reliability Analysis of Aeroengine Blades

Based on the nonlinear continuum damage model （CDM） developed by Chaboehe, a modified model for high cycle fatigue of TC4 alloy was proposed. Unsymmetrical cycle fatigue tests were conducted on rod specimens at room temperature. Then the material parameters needed in the CDM were obtained by the fatigue tests, and the stress distribution of the specimen was calculated by FE method. Compared with the linear damage model （LDM）, the dam- age results and the life prediction of the CDM show a better agreement with the test and they are more precise than the LDM. By applying the CDM developed in this study to the life prediction of aeroengine blades, it is concluded that the root is the most dangerous region of the whole blade and the shortest life is 58 211 cycles. Finally, the Cox propor- tional hazard model of survival analysis was applied to the analysis of the fatigue reliability. The Cox model takes the covariates into consideration, which include diameter, weight, mean stress and tensile strength. The result shows that the mean stress is the only factor that accelerates the fracture process.

Reliability Prediction for Internetware Applications:A Research Framework and Its Practical Use

The open and dynamic environment of Internet computing demands new software reliability technologies.How to efficiently and effectively build highly reliable Internet applications becomes a critical research problem.This paper proposes a research framework for predicting reliability of individual software entities as well as the whole Internet application.Characteristics of the Internet environment are comprehensively analyzed and several reliability prediction approaches are proposed.A prototype is implemented and practical use of the proposed framework is also demonstrated.