Exponentially Weighted Moving Average Research Articles

Warning signals for mental health problems in at‐risk young adults may be informed by momentary emotions reported by the general population: A novel application of the principles of statistical process control

AbstractStatistical process control (SPC) was recently introduced as a method for detecting person‐specific warning signals for mental ill‐health. Such warning signals occur when a person's repeatedly assessed emotions exceed a control limit. This control limit should in principle be based on the same person's emotions in a healthy period. As such data are often unavailable, this preregistered study investigated whether general population data can be used instead to estimate control limits. We used data from the HowNutsAreTheDutch study, in which adults from the general population (N = 746) rated their emotions three times a day for 1 month. Based on these data, we computed control limits according to the exponentially weighted moving average (EWMA) and Shewhart SPC methods. Next, we investigated how often young adults with versus without persistent mental health problems from the TRAILS TRANS‐ID study (N = 100)–who rated their emotions daily for 6 months–reported scores beyond these general population‐based control limits. Generally, warning signals occurred more often in young adults with persistent mental health problems compared to healthy young adults (p < 0.05). The predictive performance of SPC did not consistently improve when control limits were conditioned on individuals' age, sex, and depressive symptoms, nor differ between methods (EWMA vs. Shewhart). The different emotions that were monitored, however, affected SPC performance, so that for most settings, warning signs in feeling tired and upset were worse for detecting mental‐ill health compared to warning signs in other emotions (e.g., feeling nervous, relaxed, etc.). It follows that warning signs in individual's emotions can perhaps be monitored using relatively generic norms, derived from the general population, opening up new avenues for research and low‐threshold clinical application.

Performance deterioration of deep learning models after clinical deployment: a case study with auto-segmentation for definitive prostate cancer radiotherapy

Our study aims to explore the long-term performance patterns for deep learning (DL) models deployed in clinic and to investigate their efficacy in relation to evolving clinical practices. We conducted a retrospective study simulating the clinical implementation of our DL model involving 1328 prostate cancer patients treated between January 2006 and August 2022. We trained and validated a U-Net-based auto-segmentation model on data obtained from 2006 to 2011 and tested on data from 2012 to 2022, simulating the model’s clinical deployment starting in 2012. We visualized the trends of the model performance using exponentially weighted moving average (EMA) curves. Additionally, we performed Wilcoxon Rank Sum Test and multiple linear regression to investigate Dice similarity coefficient (DSC) variations across distinct periods and the impact of clinical factors, respectively. Initially, from 2012 to 2014, the model showed high performance in segmenting the prostate, rectum, and bladder. Post-2015, a notable decline in EMA DSC was observed for the prostate and rectum, while bladder contours remained stable. Key factors impacting the prostate contour quality included physician contouring styles, using various hydrogel spacers, CT scan slice thickness, MRI-guided contouring, and intravenous (IV) contrast (p < 0.0001, p < 0.0001, p = 0.0085, p = 0.0012, p < 0.0001, respectively). Rectum contour quality was notably influenced by factors such as slice thickness, physician contouring styles, and the use of various hydrogel spacers. The quality of the bladder contour was primarily affected by IV contrast. The deployed DL model exhibited a substantial decline in performance over time, aligning with the evolving clinical settings.

Designing signal processing for robotic olfaction based on electroantennogram

ABSTRACT In this study, we experimentally evaluated a signal-processing method for the stable use of an insect antenna as an odor sensor for a robot because of its excellent sensitivity and selectivity to odors. Changes in potential in response to an odor stimulus can be measured when the electrodes are connected to both ends of an insect antenna. This is called an electroantennogram (EAG), and is a type of physiological response. Hence, we must extract only the EAG waveform using appropriate signal processing because it has a low signal-to-noise ratio. Accordingly, we applied signal processing based on an exponentially weighted moving average (EWMA), which demonstrates excellent real-time performance, to an EAG to eliminate high-frequency noise. Moreover, we experimentally investigated the sampling rate to properly measure the EAG. From the results, the EAG-driven robot with the designed signal processing demonstrated a search performance of more than 70 % , even in odor-source localization experiments with different types of moths and environments.

Shewhart-EWMA chart for monitoring binomial data subject to shifts of random amounts

Attribute charts are used widely for monitoring binary events in manufacturing, service, and healthcare processes. While Shewhart type charts are efficacious in detecting large or sudden shifts in a nonconforming rate p, exponentially weighted moving average (EWMA) type charts tend to be more powerful for detecting smaller or gradual shifts. Since many processes can shift by various amounts, this paper proposes a combined optimized Shewhart-EWMA dual chart that combines the strengths of both charts to quickly detect shifts of any random magnitude within a defined range. The performance of this combined Shewhart-EWMA chart is optimized by allocating total detection power between the Shewhart chart element and EWMA chart element in a manner that yields the best overall effectiveness while maintaining the false alarm rate at a specified desired level. In numeric analysis under a range of settings, the detection speed of this combined Shewhart-EWMA chart is found to outperform individual Shewhart and EWMA charts, with up to 499% and 31% fewer average number of defective items respectively until process shifts of random sizes are detected. It also demonstrates enhanced overall efficacy when compared to synthetic, cumulative sum, combined Shewhart-synthetic, and combined Shewhart-CUSUM charts.

Modified exponentially weighted moving average control chart for monitoring process dispersion

The exponentially weighted moving average ( EWMA ) charts are widely used memory-type charts for monitoring small to moderate shifts in process parameters. However, the performance of the EWMA chart has been improved over time due to various modifications and enhancements. This paper proposes two modified EWMA ( M ‐ EWMA ) charts to monitor process dispersion. The upper-sided modified EWMA chart is denoted by M ‐ EWMA U , while a lower-sided modified EWMA chart is symbolized as the M ‐ EWMA L chart. Monte Carlo simulations determine the proposed schemes’ run length (RL) properties in zero- and steady-state scenarios. The proposed upper- and lower-sided M ‐ EWMA charts are compared to the upper- and lower-sided CH ‐ EWMA and SJ-EWMA charts. The comparisons reveal that the proposed M-EWMA charts have better detection ability than the upper- and lower-sided CH-EWMA, and SJ-EWMA charts, respectively. Finally, a real-life application is provided to illustrate the implementation of the upper-sided M-EWMA and CH-EWMA charts practically in both zero-state and steady-state cases.

Average Run Length Computations of Autoregressive and Moving Average Process using the Extended EWMA Procedure

In the past, the control chart served as a statistical tool for detecting process changes. The Exponentially Weighted Moving Average (EWMA) control chart is highly effective for detecting small changes. This research introduces the Extended Exponentially Weighted Moving Average (Extended EWMA) control chart for the Autoregressive and Moving average process with order p = 1 and q = 1 (ARMA(1,1)) The Extended EWMA control chart incorporates two smoothing parameters ( λ1 and λ2 ) derived from the EWMA control chart. A comparative analysis of the performance of the EWMA control chart. The Average Run Length (ARL) value as determined by the explicit formulas in this research, serves as a metric for evaluating the performance of the Extended EWMA control chart and the EWMA control chart. The Numerical Integral Equation (NIE) method is used to verify the accuracy of the ARL for the explicit formulas of the two control charts which has not been before discovered. The effectiveness of control charts can also be evaluated by analyzing SDRL, ARL, MRL, RMI, AEQL, and PCI values as metrics for various design parameter values. After analyzing the results of the ARL and all five performance meters, it was determined that the Extended EWMA control chart is better than the EWMA control chart at all shift sizes of process changes. Finally, the assessment of the ARMA process is being conducted to evaluate the ARL using a dataset on PM2.5 dust levels in Bangkok, Thailand during January and February of 2024.

Approximating the ARL to Monitor Small Shifts in the Mean of an AR Fractionally Integrated with an exogenous variable Process Running on an EWMAControl Chart

Control charts are used to monitor processes and detect changes in a given control scheme. The Exponential Weighted Moving Average (EWMA) control chart is a well-recognized control chart used to detect small changes in parameters. The efficiency of the chart studied is usually achieved using ARL. Approximating ARL using the Gauss-Legendre quadrature method, also known as NIE,. This approach is used to evaluate the ARL of developments, such as explicit formulas because it provides a robust way to validate their validity and accuracy. Moreover, it evaluates the performance of control charts for time series under exponential white noise. Exponential white noise is obtained from a long-memory fractionally integrated AR with exogenous variables or the long-memory ARFIX process. Under the long-memory ARFIX model, the proposed technique compares the control chart's performance to an explicit formula using the criterion of percentage accuracy. The results of the comprehensive numerical study include investigations into a wide range of out-of-control processes and situations. Specifically, the results from the accuracy percentage in all cases are more than 95%, which means that the proposed technique is accurate and completely consistent with the well-defined explicit formula. Therefore, it is recommended that it be used in this situation. There are examples from real data that were found to be consistent with the research results.

Enhancing Public Health Surveillance: Outbreak Detection Algorithms Deployed for Syndromic Surveillance During Arbaeenia Mass Gatherings in Iraq.

Large gatherings often involve extended and intimate contact among individuals, creating environments conducive to the spread of infectious diseases. Despite this, there is limited research utilizing outbreak detection algorithms to analyze real syndrome data from such events. This study sought to address this gap by examining the implementation and efficacy of outbreak detection algorithms for syndromic surveillance during mass gatherings in Iraq. For the study, 10 data collectors conducted field data collection over 10 days from August 25, 2023, to September 3, 2023. Data were gathered from 10 healthcare clinics situated along Ya Hussein Road, a major route from Najaf to Karbala in Iraq. Various outbreak detection algorithms, such as moving average, cumulative sum, and exponentially weighted moving average, were applied to analyze the reported syndromes. During the 10 days from August 25, 2023, to September 3, 2023, 12202 pilgrims visited 10 health clinics along a route in Iraq. Most pilgrims were between 20 and 59 years old (77.4%, n=9444), with more than half being foreigners (58.1%, n=7092). Among the pilgrims, 40.5% (n=4938) exhibited syndromes, with influenza-like illness (ILI) being the most common (48.8%, n=2411). Other prevalent syndromes included food poisoning (21.2%, n=1048), heatstroke (17.7%, n=875), febrile rash (9.0%, n=446), and gastroenteritis (3.2%, n=158). The cumulative sum (CUSUM) algorithm was more effective than exponentially weighted moving average (EWMA) and moving average (MA) algorithms for detecting small shifts. Effective public health surveillance systems are crucial during mass gatherings to swiftly identify and address emerging health risks. Utilizing advanced algorithms and real-time data analysis can empower authorities to improve their readiness and response capacity, thereby ensuring the protection of public health during these gatherings.

An algorithm for modelling escalator fixed loss energy for PHM and sustainable energy usage

Fixed loss energy of escalators denotes the energy consumption of an unloaded escalator. Fixed loss energy fluctuates over time indicating varying operation conditions. We propose an algorithm (named the Optimization approach) to compute daily fixed loss energy based on energy consumption sensor data and validate our algorithm by comparing it to two benchmark methods (named the Classical and Engineering approaches) using a set of experimental data. Our evaluation includes a discussion of the advantages and disadvantages of each approach, ultimately recommending our method due to its nearly 20 times smaller bias. Our optimization approach shows significant improvement (p-value=0.022): error declines from 2.87 for the classical approach to -0.11 for our approach. To further highlight the practical application within Prognostic Health Management (PHM), we design an Exponential Weighted Moving Average (EWMA) chart for monitoring the fixed loss over time and illustrate the potential of this approach in reducing energy usage and costs associated with escalator operation. We estimate that 5 Wh of energy every minute can be saved, which roughly equals 10% of the fixed loss energy of the escalator in our example. This study enables us to support health status assessment of escalators, and provides valuable insights into maintenance effectiveness, as well as identifying excessive energy consumption.

Combined Shewhart–EWMA and Shewhart–CUSUM monitoring schemes for time between events

AbstractTo improve the detecting abilities of upward and downward parameter changes in high‐quality process, two combined schemes by integrating memory‐type, that is, exponentially weighted moving average (EWMA) or cumulative sum (CUSUM), and memoryless, that is, Shewhart, are proposed for monitoring the time between events (TBE), which is modeled as an exponential distributed variable. A Monte Carlo simulation method is employed to obtain the Run Length () properties, that is average run length (), of the proposed schemes for different parameter settings. The nearly optimal monitoring parameter combinations under different change size are obtained by minimizing the out‐of‐control with the satisfied in‐control . Using the designed parameters, the performance of the proposed monitoring schemes is compared with the existing EWMA and CUSUM TBE. The results show that the proposed combined Shewhart–EWMA or Shewhart–CUSUM TBE generally perform better than the corresponding EWMA or CUSUM TBE for large changes and they also show better performance than the Shewhart TBE for small changes. Finally, a real dataset of organic light‐emitting diode (OLED) failure time from Sumsung company is employed to indicate the usage and implementation of combined TBE schemes.

Control chart for detecting the scale parameter of the zero‐inflated Poisson model

AbstractWhen monitoring risk in public health, count data commonly exhibit an excessive number of zero, and the zero‐inflated Poisson (ZIP) model is often used to fit this type of data. Most previous methods for monitoring of the ZIP model have focused on the changes in the location parameter and the existence of the scale parameter and usually assumed that the scale parameter is zero in the H0 stage. However, in an objective environment, data often have certain fluctuations, meaning that the scale parameter always exists. Therefore, it is more meaningful to monitor the changes in the scale parameter on top of the predefined baseline than to monitor its existence. In this study, we derive a score test statistic based on the generalized Henderson's joint likelihood function, construct a risk‐adjusted exponentially weighted moving average (EWMA) control chart to monitor the variability of the random effects variance component in the ZIP mixed‐effects model. And the convergence property of the score test statistic is proved through derivation, which shows that the new method has theoretical reliability. The simulation results Indicate that when the scale parameter has different predefined baselines and different variation amplitudes, the proposed method is more effective than the existing RA‐ZIP and PR‐ZIP control charts. In addition, the proposed method is applied to real data from a Hong Kong hospital for online influenza surveillance to demonstrate its practicability.

A critique on the use of the belief statistic for process monitoring

AbstractRecently, several control charts have been proposed in the statistical process monitoring literature based on a belief statistic. In this article we compare the zero‐state average run‐lengths and conditional expected delay profiles of the exponentially weighted moving average (EWMA) and belief statistic‐based (BSB) charts. We show that an ordinary EWMA chart is far better than the BSB chart when detecting delayed shifts in the mean of a normally distributed process.

Developing Bayesian EWMA chart for change detection in the shape parameter of Inverse Gaussian process.

Bayesian Control charts are emerging as the most efficient statistical tools for monitoring manufacturing processes and providing effective control over process variability. The Bayesian approach is particularly suitable for addressing parametric uncertainty in the manufacturing industry. In this study, we determine the monitoring threshold for the shape parameter of the Inverse Gaussian distribution (IGD) and design different exponentially-weighted-moving-average (EWMA) control charts based on different loss functions (LFs). The impact of hyperparameters is investigated on Bayes estimates (BEs) and posterior risks (PRs). The performance measures such as average run length (ARL), standard deviation of run length (SDRL), and median of run length (MRL) are employed to evaluate the suggested approach. The designed Bayesian charts are evaluated for different settings of smoothing constant of the EWMA chart, different sample sizes, and pre-specified false alarm rates. The simulative study demonstrates the effectiveness of the suggested Bayesian method-based EWMA charts as compared to the conventional classical setup-based EWMA charts. The proposed techniques of EWMA charts are highly efficient in detecting shifts in the shape parameter and outperform their classical counterpart in detecting faults quickly. The proposed technique is also applied to real-data case studies from the aerospace manufacturing industry. The quality characteristic of interest was selected as the monthly industrial production index of aircraft from January 1980 to December 2022. The real-data-based findings also validate the conclusions based on the simulative results.

Sine hunter prey optimization enabled deep residual network for diabetes mellitus detection using tongue image

Background: Many people suffer from Diabetes Mellitus (DM), a disease caused by high blood glucose levels. In real-time, many methods are implemented to diagnose DM to obtain a good accuracy level, but those methods remain costlier. Objective: To develop a method for DM detection with good accuracy and minimum cost. Materials and methods: In this research, DM is detected using tongue image based on DL model, named Deep Residual Network (DRN) that is trained by proposed Sine Hunter Prey Optimization (SHPO). Here, an adaptive median filter is used for the pre-processing phase, and image segmentation is done using ResUNet++, which is trained by Exponential Anti Corona Virus Optimization (ExpACVO). Here, ExpACVO integrates Anti Corona Virus Optimization (ACVO) and Exponential Weighted Moving Average (EWMA). Further, image augmentation and appropriate feature extraction stages are carried out, leading to DM detection by DRN. Moreover, SHPO is formed by combining the Sine Cosine Algorithm (SCA) and Hunter Prey Optimization (HPO). The performance of the proposed method is analyzed using the Tongue image dataset and the Diabetic images dataset. Results: The performance of SHPO_DRN is found using four evaluation metrics: accuracy, sensitivity, specificity, and f-measure. Here, these metrics exhibit superior performance with high-range values of 0.961, 0.970, 0.948, and 0.961. Conclusion: The proposed method detects the DM at earlier stages with a good accuracy.

Two Bayesian approaches of monitoring mean of Gaussian process using Bayes factor

AbstractThis paper develops two novel process monitoring schemes for the mean of a Gaussian process: the Bayes factor (BF) and the improved Bayes factor (IBF) schemes. Conjugate priors are used to construct the plotting statistics. The performance of the proposed schemes is evaluated in terms of average run length (ARL), standard deviation of run length (SDRL), and several percentiles, and these performance metrics across different hyper‐parameters and various sample sizes are evaluated via Monte Carlo simulations. Both zero‐state and steady‐state out‐of‐control (OOC) performances are investigated comprehensively. The simulation results show that the IBF scheme outperforms the existing Bayesian exponentially weighted moving average (EWMA) schemes under different loss functions in zero‐state. In steady‐state conditions, the IBF scheme outperforms for small shifts. Finally, we present two examples to illustrate the practical application of the proposed schemes.

Calibrated EWMA estimators for time-scaled surveys with diverse applications

Recently, several memory-type mean estimators (including ratio, product, and logarithmic) have been developed. These estimators rely on exponentially weighted moving average (EWMA), which incorporate both historical and present sample data. In this article, we propose EWMA type calibrated estimators under single and double stratified random sampling (StRS). Because calibration method enhances the estimates by modifying the stratification weight, taking advantage of supplementary information. To evaluate the performance of estimators, various real-world time-scaled data sets pertaining to stock market and weather are taken into account. Additionally, we also conduct a simulation study using a bivariate symmetric data set. The numerical results show the superiority of proposed estimators (y¯TM,y¯TaM) over the adapted ones (y¯PM,y¯PaM).

Risk adjusted EWMA control chart based on support vector machine with application to cardiac surgery data

In the current study, we demonstrate the use of a quality framework to review the process for improving the quality and safety of the patient in the health care department. The researchers paid attention to assessing the performance of the health care service, where the data is usually heterogeneous to patient’s health conditions. In our study, the support vector machine (SVM) regression model is used to handle the challenge of adjusting the risk factors attached to the patients. Further, the design of exponentially weighted moving average (EWMA) control charts is proposed based on the residuals obtained through SVM regression model. Analyzing real cardiac surgery patient data, we employed the SVM method to gauge patient condition. The resulting SVM-EWMA chart, fashioned via SVM modeling, revealed superior shift detection capabilities and demonstrated enhanced efficacy compared to the risk-adjusted EWMA control chart.

EWMA control charts based on robust estimators: A powerful tool for monitoring a process with a non-normal distribution

This paper presents a study on using memory-type control charts to detect small to moderate shifts in the process mean, particularly when the underlying distribution deviates from normal assumptions. To address this, robust estimators based on modified maximum likelihood estimation (MMLE) and Lloyd’s BLUE estimators using generalized least square estimation (GLSE) are proposed. The proposed estimators utilize auxiliary information to enhance sensitivity against abrupt changes in the mean. Additionally, they play a crucial role in refining the precision of the estimated mean, contributing to a more accurate and nuanced estimation of the true mean. The efficiency and robustness of these estimators are compared with conventional ones from the literature. The proposed robust estimators are integrated into the design of the exponentially weighted moving average (EWMA) control chart. Performance evaluation is based on average run lengths (ARLs), standard deviation, and percentiles of the run lengths. A real-life example illustrates the significance of the proposed estimators over conventional ones for non-normal distributions. The results demonstrate that the EWMA control chart with proposed estimators outperforms competitors in terms of ARLs, SDRLs, and percentiles. This study offers insights into using robust estimators to enhance memory-type control charts' performance for non-normal distributions.

Monitoring method and application of transition process with nonstationary conditions based on stability factor partitioning and RSFA

It is common for the working conditions to change with time in actual industrial processes. However, the transition modes of complex industrial processes under different working conditions often have various degrees of dynamic nonstationarity, which makes the traditional process monitoring model based on the stationarity assumption ineffective. In this paper, a Recursive Slow Feature Analysis method based on Stability Factor Partitioning (SFP-RSFA) is proposed for fine process monitoring of transition modes under dynamic nonstationarity characteristics. First, we calculate the stability factor according to the different stationarity characteristics of the production process variables. Then, K-means clustering is carried out according to the stability factor of each variable, and the stability factor of the cluster center is mapped to the interval [0,1] as the smoothing coefficient of the exponential weighted moving average (EWMA), which is applied to each data subblock respectively to highlight the steady-state and dynamic characteristics of the monitoring data subblock. In the online monitoring stage, the monitored data are fed into the subblock recursive slow feature analysis (RSFA) monitoring model. Finally, a comprehensive statistic method is proposed to integrate the subblock monitoring statistics. The Tennessee Eastman (TE) process and actual cement clinker production process were tested and compared with existing RPCA, RCA and RSFA methods. The effectiveness and superiority of the proposed method in the problem of nonstationary transition mode process monitoring are verified.

A distribution-free Max-EWMA scheme for multi-aspect process monitoring with industrial applications

In Statistical Process Monitoring, Distribution-free charting schemes play a vital role in Industry 4.0 for their superior effectiveness compared to parametric methods, especially in cases where the underlying process distribution is not well-defined or poses challenges in estimation. Traditional distribution-free charting schemes focus on surveilling only one process aspect: the location or scale parameter. The past two decades have also witnessed a rapid expansion of research on joint monitoring of the two aspects, location and scale, using combined bi-aspect charting schemes. However, there is a notable scarcity of methods that addressing complex industrial scenarios where multiple parameters (like location, scale, and shape) need simultaneous monitoring with high sensitivity and specificity. Our research is motivated by this challenge, and this paper describes a distribution-free tri-aspect monitoring procedure called the Maximum Exponentially Weighted Moving Average (Max-EWMA) scheme This scheme can effectively monitor shifts in three process characteristics, encompassing aspects related to process location, scale, and shape parameters. We utilize weighted versions of well-known statistical measures such as the Wilcoxon, Ansari-Bradley, and Savage-type statistics to construct the plotting statistics. An essential feature of our proposed Max-EWMA chart is its robustness under various continuous distributions, ensuring reliable performance in in-control situations. Competing schemes are compared via intensive computing techniques based on Monte Carlo using the Median Run Length metric, which evaluates the effectiveness of the proposed scheme. Finally, our proposed schemes are illustrated with two examples. Some concluding remarks and limitations of the study are noted.