Search results

When a process experiences an out-of-control condition, identification of the change point is capable of leading practitioners to an effective root cause analysis. The change point addresses the time when a special cause(s) manifests itself into the process. In the statistical process monitoring when the chart signals an out-of-control condition, the change point analysis is an important step for the root cause analysis of the process. This paper attempts to propose a model approaching the artificial neural network to identify the change point of a multistage process with cascade property in the case that the process is modeled properly by a simple linear profile.

In practice, many processes can be modeled by a functional relationship rather than a single random variable or a random vector. This approach of modeling is referred to as the profile in the statistical process control literature. In this paper, two models based on multilayer perceptron (MLP) and convolutional neural network (CNN) approaches are proposed for identifying the change point of the profile of a multistage process.

The capability of the proposed models are evaluated and compared using several numerical scenarios. The numerical analysis of the proposed neural networks indicates that the two proposed models are capable of identifying the change point in different scenarios effectively. The comparative sensitivity analysis shows that the capability of the proposed convolutional network is superior compared to MLP network.

To the best of the authors' knowledge, this is the first time that: (1) A model is proposed to identify the change point of the profile of a multistage process. (2) A convolutional neural network is modeled for identifying the change point of an out-of-control condition.

The purpose of this paper is to develop a new software reliability growth model considering different fault distribution function before and after the change point.

In this paper, the authors have developed a framework to incorporate change-point in developing a hybrid software reliability growth model by considering different distribution functions before and after the change point.

Numerical illustration suggests that the proposed model gives better results in comparison to the existing models.

The existing literature on change point-based software reliability growth model assumes that the fault correction trend before and after the change is governed by the same distribution. This seems impractical as after the change in the testing environment, the trend of fault detection or correction may not follow the same trend; hence, the assumption of same distribution function may fail to predict the potential number of faults. The modelling framework assumes different distributions before and after change point in developing a software reliability growth model.

The purpose of this paper is to propose two control charts to monitor multi-attribute processes and then a maximum likelihood estimator for the change point of the parameter vector (process fraction non-conforming) of multivariate binomial processes.

The performance of the proposed estimator is evaluated for both control charts using some simulation experiments. At the end, the applicability of the proposed method is illustrated using a real case.

The proposed estimator provides accurate and useful estimation of the change point for almost all of the shift magnitudes, regardless of the process dimension. Moreover, based on the results obtained the estimator is robust with regard to different correlation values.

To the best of authors’ knowledge, there are no work available in the literature to estimate the change-point of multivariate binomial processes.

Almost everything around us is the output of software-driven machines or working with software. Software firms are working hard to meet the user’s requirements. But developing a fault-free software is not possible. Also due to market competition, firms do not want to delay their software release. But early release software comes with the problem of user reporting more failures during operations due to more number of faults lying in it. To overcome the above situation, software firms these days are releasing software with an adequate amount of testing instead of delaying the release to develop reliable software and releasing software patches post release to make the software more reliable. The paper aims to discuss these issues.

The authors have developed a generalized framework by assuming that testing continues beyond software release to determine the time to release and stop testing of software. As the testing team is always not skilled, hence, the rate of detection correction of faults during testing may change over time. Also, they may commit an error during software development, hence increasing the number of faults. Therefore, the authors have to consider these two factors as well in our proposed model. Further, the authors have done sensitivity analysis based on the cost-modeling parameters to check and analyze their impact on the software testing and release policy.

From the proposed model, the authors found that it is better to release early and continue testing in the post-release phase. By using this model, firms can get the benefits of early release, and at the same time, users get the benefit of post-release software reliability assurance.

The authors are proposing a generalized model for software scheduling.

Despite being a widely used management technique, cumulative sum (CUSUM) analysis remains almost unheard of in professional sport. To address this, CUSUM analysis of soccer match data from the English Premier League (EPL) was performed. The primary objective of the study was to evaluate CUSUM as a tool for assessing “on-field” team performance. As a secondary objective, the association between managerial change and team performance was evaluated.

CUSUM was applied retrospectively to goal difference data for six EPL teams (Arsenal, Chelsea, Everton, Liverpool, Manchester United and Tottenham) over 23 consecutive seasons from 1995 to 2018. This was supplemented with change point analysis to identify structural changes in mean goal difference. Succession was evaluated by mapping historical managerial changes onto the CUSUM plots for the respective clubs.

CUSUM analysis revealed the presence of structural changes in four clubs. Two structural change points were identified for both Chelsea and Everton, one for Manchester United and Tottenham and none for Arsenal and Liverpool. Relatively few managerial changes coincided temporally with structural changes in “on-field” performance, with most appointments having minimal impact on long-term team performance. Other factors (e.g. changes in ownership) appear to have been influential.

The study was limited by the fact that only successful teams were investigated.

CUSUM analysis appears to have potential as a tool for executive decision-makers to evaluate performance outcomes in professional soccer.

The study is the first of its kind to use CUSUM analysis to evaluate team performance in professional soccer.

The purpose of this paper is to investigate changes of China’s agri-food exports to Germany caused by China’s accession to WTO and the global financial crisis in a quantitative way. The paper aims to detect structural breaks and compare differences before and after the change points.

The structural breaks detection procedures in this paper can be applied to find out two different types of change points, i.e. in the middle and at the end of one time series. Then time series and regression models are used to compare differences of trade relationship before and after the detected change points. The methods can be employed in any economic series and work well in practice.

The results indicate that structural breaks in 2002 and 2009 are caused by China’s accession to WTO and the financial crisis. Time series and regression models show that the development of China’s exports to Germany in agri-food products has different features in different sub-periods. Before 1999, there is no significant relationship between China’s exports to Germany and Germany’s imports from the world. Between 2002 and 2008 the former depends on the latter very strongly, and China’s exports to Germany developed quickly and stably. It decreased, however suddenly in 2009, caused by the great reduction of Germany’s imports from the world in that year. But China’s market share in Germany still had a small gain. Analysis of two categories in agri-food trade also leads to similar conclusions. Comparing the two events we see rather different patterns even if they both indicate structural breaks in the development of China’s agri-food exports to Germany.

This paper partly originally proposes two statistical algorithms for detecting different kinds of structural breaks in the middle part and at the end of a short-time series, respectively.

This paper aims to develop a parsimonious and innovative model that captures the dynamics of new product diffusion in the recent high-technology markets and thus assist both academicians and practitioners who are eager to understand the diffusion phenomena. Accordingly, this study develops a novel diffusion model to forecast the demand by centering on the dynamic state of the product’s adoption rate. The proposed study also integrates the consumer’s psychological point of view on price change and goodwill of the innovation in the diffusion process.

In this study, a two-dimensional distribution function has been derived using Cobb–Douglas’s production function to combine the effect of price change and continuation time (goodwill) of the technology in the market. Focused on the realistic scenario of sales growth, the model also assimilates the time-to-time variation in the adoption rate (hazard rate) of the innovation owing to companies changing marketing and pricing strategies. The time-instance upon which the adoption rate alters is termed as change-point.

For validation purpose, the developed model is fitted on the actual sales and price data set of dynamic random access memory (DRAM) semiconductors, liquid crystal display (LCD) monitors and room air-conditioners using non-linear least squares estimation procedure. The results indicate that the proposed model has better forecasting efficiency than the conventional diffusion models.

The developed model is intrinsically restricted to a single generation diffusion process. However, technological innovations appear in generations. Therefore, this study also yields additional plausible directions for future analysis by extending the diffusion process in a multi-generational environment.

This study aims to assist marketing managers in determining the long-term performance of the technology innovation and examine the influence of fluctuating price on product demand. Besides, it also incorporates the dynamic tendency of adoption rate in modeling the diffusion process of technological innovations. This will support the managers in understanding the practical implications of different marketing and promotional strategies on the adoption rate.

This is the first attempt to study the value-based diffusion model that includes key interactions between goodwill of the innovation, price dynamics and change-point for anticipating the sales behavior of technological products.

The stock market price time series can be divided into two processes: continuously rising and continuously falling. The authors can effectively prevent the stock market from crashing by accurately estimating the risk on continuously rising returns (CRR) and continuously falling returns (CFR).

The authors add an exogenous variable into Log-autoregressive conditional duration (Log-ACD) model, and then apply our extended Log-ACD model and Archimedean copula to estimate the marginal distribution and conditional distribution of CRR and CFR. Plus, the authors analyze the conditional value at risk (CVaR) and present back-test results of the CVaR. The back-test shows that our proposed risk estimation method has a good estimation power for the risk of the CRR and CFR, especially the downside risk. In addition, the authors detect whether the dependent structure between the CRR and CFR changes using the change point test method.

The empirical results indicate that there is no change point here, suggesting that the results on the dependent structure and risk analysis mentioned above are stable. Therefore, major financial events will not affect the dependent structure here. This is consistent with the point that the CRR and CFR can be analyzed to obtain the trend of stock returns from a more macro perspective than daily stock returns scholars usually study.

The risk estimation method of this paper is of great significance in understanding stock market risk and can provide corresponding valuable information for investment advisors and public policy regulators.

The authors defined a new stock returns, CRR and CFR, since it is difficult to analyze and predict the trend of stock returns according to daily stock returns because of the small autocorrelation among daily stock returns.

Visual inspection and human judgement form the cornerstone of daily operations and maintenance (O&M) services activities carried out by facility managers nowadays. Recent advances in technologies such as building information modelling (BIM), distributed sensor networks, augmented reality (AR) technologies and digital twins present an immense opportunity to radically improve the way daily O&M is conducted. This paper aims to describe the development of an AR-supported automated environmental anomaly detection and fault isolation method to assist facility managers in addressing problems that affect building occupants’ thermal comfort.

The developed system focusses on the detection of environmental anomalies related to the thermal comfort of occupants within a building. The performance of three anomaly detection algorithms in terms of their ability to detect indoor temperature anomalies is compared. Based on the fault tree analysis (FTA), a decision-making tree is developed to assist facility management (FM) professionals in identifying corresponding failed assets according to the detected anomalous symptoms. The AR system facilitates easy maintenance by highlighting the failed assets hidden behind walls/ceilings on site to the maintenance personnel. The system can thus provide enhanced support to facility managers in their daily O&M activities such as inspection, recording, communication and verification.

Taking the indoor temperature inspection as an example, the case study demonstrates that the O&M management process can be improved using the proposed AR-enhanced inspection system. Comparative analysis of different anomaly detection algorithms reveals that the binary segmentation-based change point detection is effective and efficient in identifying temperature anomalies. The decision-making tree supported by FTA helps formalise the linkage between temperature issues and the corresponding failed assets. Finally, the AR-based model enhanced the maintenance process by visualising and highlighting the hidden failed assets to the maintenance personnel on site.

The originality lies in bringing together the advances in augmented reality, digital twins and data-driven decision-making to support the daily O&M management activities. In particular, the paper presents a novel binary segmentation-based change point detection for identifying temperature anomalous symptoms, a decision-making tree for matching the symptoms to the failed assets, and an AR system for visualising those assets with related information.

Explains that the shifts of a process may be classified into a set of modes (or classifications), each of which is incurred by an assignable cause. Presents an algorithm to determine the process shift mode and estimate the run length when an out‐of‐control status is signalled by the x‐ or s chart in statistical process control. The information regarding the process shift mode and run length is very useful for diagnosing the assignable cause correctly and promptly. The algorithm includes two stages. First, the process shift modes are established using the sample data acquired during an explorative run. Afterwards, whenever an out‐of‐control case is detected, Bayes’ rule is employed to determine the active process shift mode and estimate the run length. In simulation tests, the proposed algorithm attains a fairly high probability (around 0.85) of correctly determining the active process shift mode and estimating the run length.