Dynamic control limits can be useful in designing control charts, especially when sample sizes, risk scores, or other covariate values change over time. Computer simulation can be used to control the conditional false alarm rate and thus the in‐control run length properties. We show that this approach can be useful in designing adaptive exponentially weighted moving average (AEWMA) control charts for which the control chart smoothing parameter at a given time point depends on the observed value at that time point. We use AEWMA charts as examples, but the approach can be applied to the adaptive cumulative sum (CUSUM) chart and other types of adaptive charts.

We review the literature on statistical process monitoring methods based on neutrosophic principles. We question some of the underlying assumptions and raise important questions about these and other neutrosophic statistical methods that need to be addressed before the methodology could be taken seriously.

An integral part of the design of control charts, including the multivariate exponentially weighted moving average (MEWMA) control chart, is the determination of the appropriate control limits for prospective monitoring. Methods using Markov chain analyses, integral equations, and simulation have been proposed to determine the MEWMA chart limits when the limits are based on a specified in-control average run length (ARL) value. A drawback of the usual approach is that the conditional false alarm rate (CFAR) for these charts varies over time in what might be in an unexpected and undesirable way. We define the CFAR as the probability of a false alarm given no previous false alarm. We do not condition on the results of a Phase I sample, as done by others, in studies of the effect of estimation error on control chart performance. We propose the use of dynamic probability control limits (DPCLs) to keep the CFAR constant over time at a specified value. The CFAR at any time, however, could be controlled to be any specified value using our approach. Using simulation, we determine the DPCLs for the MEWMA control chart being used to monitor the mean vector with an assumed known variance-covariance matrix. We consider cases where the sample size is both fixed and time-varying. For varying sample sizes, the DPCLs adapt automatically to any change in the sample size distribution. In all cases, the CFAR is held closely to a fixed value and the resulting in-control run length performance follows closely to that of the geometric distribution.

In this article, the panelists broadly discuss the definition of network monitoring, and how it may be similar to or different from network surveillance and network change-point detection. The discussion uncovers ambiguity and contradictions associated with these terms and we argue that this lack of clarity is detrimental to the field. The panelists also describe existing and emerging applications of network monitoring, which serves to illustrate the wide applicability of the tools and research associated with the field.

The conditional false alarm rate (CFAR) at a particular time is the probability of a false alarm for an assumed in-control process at that time conditional on no previous false alarm. Only the Shewhart control chart designed with known in-control parameters, or conditioned on the estimated parameters, has a constant conditional false alarm rate. Other types of charts, however, can have their control limits determined in order to have any desired pattern of CFARs. The important advantage of the use of this CFAR metric is when sample sizes, population sizes or other covariate information affecting chart performance vary over time. In these cases, the control limit at a particular time can be obtained through control of the CFAR value after the corresponding covariate value is known. This allows one to control the in-control performance of the chart without the need to model or forecast the covariate values. The approach is illustrated using the risk-adjusted Bernoulli cumulative sum (CUSUM) chart.

The study and use of network monitoring methodology is informed by its need in government, industry, and technology. Here, the panelists discuss the broader impacts of network monitoring in these sectors, how the use and development of new methods is influenced by these institutions, and what challenges need to be addressed in the next 5 to 10 years. There is a strong consensus that these sectors each play an important role in the innovation of network monitoring techniques. Applications to cyber security, transportation, infectious disease monitoring, engineering, and artificial intelligence are discussed.

Research in network monitoring spans a large and growing number of disciplines, including mathematics, physics, computer science, and statistics. Here, the panelists discuss the advantages and disadvantages of the interdisciplinary nature of the area. It is largely agreed that integrating expertise from many disciplines drives innovation in network monitoring development, but several notable barriers are discussed that limit the area's full potential.

Traditional statistical process monitoring (SPM) provides a useful starting point for framing and solving network monitoring problems. In this paper the panelists discuss similarities and differences between the two fields and they describe many challenges and open problems in contemporary network monitoring research. The panelists also discuss potential outlets and avenues for disseminating such research.

Creating an interactive, accurate, and low-latency big data visualisation is challenging due to the volume, variety, and velocity of the data. Visualisation options range from visualising the entire big dataset, which could take a long time and be taxing to the system, to visualising a small subset of the dataset, which could be fast and less taxing to the system but could also lead to a less-beneficial visualisation as a result of information loss. The main research questions investigated by this work are what effect sampling has on visualisation insight and how to provide guidance to users in navigating this trade-off. To investigate these issues, we study an initial case of simple estimation tasks on histogram visualisations of sampled big data, in hopes that these results may generalise. Leveraging sampling, we generate subsets of large datasets and create visualisations for a crowd-sourced study involving a simple cognitive visualisation task. Using the results of this study, we quantify insight, sampling, visualisation, and perception error in comparison to the full dataset. We use these results to model the relationship between sample size and insight error, and we propose the use of our model to guide big data visualisation sampling.

Timely detection of anomalous events in networks, particularly social networks, is a problem of increasing interest and relevance. A variety of methods have been proposed for monitoring such networks, including the window‐based scan method proposed by a previous study. However, research assessing the performance of this and other methods has been sparse. In this article, we use simulated social network structures to study the performance of the Priebe et al method. The detection power is high only when more than half of the social network experiences anomalous behavior or if the anomalous behavior is extreme. Both can be represented by high signal‐to‐noise ratios in the network. More precisely, Priebe's scan method performs well when the signal‐to‐noise ratio is above 20. Simulation studies are used to show that an improved detection rate and shortened monitoring delays can be achieved by lagging the moving window used for standardization, lowering the signaling threshold, and using shorter moving windows at the initial stage of monitoring. We suggest a community detection method to be used after an anomalous event has been identified to help determine the subnetwork associated with this anomalous behavior.

Social networks have become ubiquitous in modern society, which makes social network monitoring a research area of significant practical importance. Social network data consist of social interactions between pairs of individuals that are temporally aggregated over a certain interval of time, and the level of such temporal aggregation can have substantial impact on social network monitoring. There have been several studies on the effect of temporal aggregation in the process monitoring literature, but no studies on the effect of temporal aggregation in social network monitoring. We use the degree corrected stochastic block model (DCSBM) to simulate social networks and network anomalies and analyze these networks in the context of both count and binary network data. In conjunction with this model, we use the Priebe scan method as the monitoring method. We demonstrate that temporal aggregation at high levels leads to a considerable decrease in the ability to detect an anomaly within a specified time period. Moreover, converting social network communication data from counts to binary indicators can result in a significant loss of information, hindering detection performance. Aggregation at an appropriate level with count data, however, can amplify the anomalous signal generated by network anomalies and improve detection performance. Our results provide both insights on the practical effects of temporal aggregation and a framework for the study of other combinations of network models, surveillance methods, and types of anomalies.

The integrity of Phase II control charting depends on the accuracy of Phase I estimation. Studies have shown that extremely large sample sizes are needed in Phase I to ensure that performance of control charts with estimated in‐control parameters is comparable with the performance of charts with known parameters. The sample size recommendations can be impractical for attribute control charts. In this article, the in‐control performance of the c ‐chart with an estimated in‐control average number of non‐conforming items is assessed. We show that the sampling variability associated with estimation results in a high percentage of control charts with in‐control average run lengths well below that of corresponding control charts with known parameters. This sampling variability can be described as between‐practitioner variability. To overcome the variability in performance, a c ‐chart with bootstrapped control limits is recommended. A simulation study reveals that these adjusted bootstrapped control limits improve the conditional average run length performance of the c ‐chart by controlling the proportion of charts with in‐control average run length performance below a given value. The out‐of‐control performance of the c ‐chart with adjusted limits is also discussed. Copyright © 2016 John Wiley & Sons, Ltd.

A growing number of applications involve monitoring with rare event data. The event of interest could be, for example, a nonconforming manufactured item, a congenital malformation, or an industrial accident. The most common approaches for monitoring such processes involve using an exponential distribution to model the time between the events or using a Bernoulli distribution to model whether or not each opportunity for the event results in its occurrence. The use of a sequence of independent Bernoulli random variables leads to a geometric distribution for the number of non-occurrences between the occurrences of the rare events. One surveillance method is to use a power transformation on the exponential or geometric observations to achieve approximate normality of the in-control distribution and then use a standard individuals control chart. We add to the argument that use of this approach is very counterproductive and cover some alternative approaches. We discuss the choice of appropriate performance metrics. The strong adverse effect of Phase I parameter estimation on Phase II performance of various charts is then summarized. In addition, the important practical issue of the effect of aggregation of counts over time, some generalizations of standard methods, and some promising research ideas are discussed.

Companies routinely perform life tests on their products. Each of these life tests typically involves testing several units simultaneously with interest in the times to failure. Two aspects often associated with lifetime data that make the development of a control-charting procedure more demanding are that the data tend to be nonnormally distributed and censored. In this paper, one-sided lower and upper likelihood-ratio-based cumulative sum (CUSUM) control charting procedures are developed for Type I right-censored Weibull lifetime data to monitor changes in the scale parameter, also known as the characteristic life, for a fixed value of the Weibull shape parameter. Because a decrease in the characteristic life indicates a decrease in the mean lifetime of a product, a one-sided lower CUSUM chart is the main focus. We illustrate the development and implementation of the chart and evaluate its properties through a simulation study. The proposed CUSUM chart is compared with an exponentially weighted moving-average (EWMA) chart using the steady-state average run length (ARL) performance. The CUSUM chart is shown to perform better than the EWMA chart in detecting shifts for which it is designed.