A setwise EWMA scheme for monitoring high-dimensional datastreams

2019 ◽  
Vol 09 (02) ◽  
pp. 2050004
Author(s):  
Long Feng ◽  
Haojie Ren ◽  
Changliang Zou
Keyword(s):  
Moving Average ◽  
Reference Sample ◽  
Divide And Conquer ◽  
High Dimensional ◽  
Dependence Structure ◽  
Test Statistic ◽  
Asymptotic Results ◽  
Statistical Process ◽  
Low Dimensional ◽  
Weighted Moving Average

The monitoring of high-dimensional data streams has become increasingly important for real-time detection of abnormal activities in many statistical process control (SPC) applications. Although the multivariate SPC has been extensively studied in the literature, the challenges associated with designing a practical monitoring scheme for high-dimensional processes when between-streams correlation exists are yet to be addressed well. Classical [Formula: see text]-test-based schemes do not work well because the contamination bias in estimating the covariance matrix grows rapidly with the increase of dimension. We propose a test statistic which is based on the “divide-and-conquer” strategy, and integrate this statistic into the multivariate exponentially weighted moving average charting scheme for Phase II process monitoring. The key idea is to calculate the [Formula: see text] statistics on low-dimensional sub-vectors and to combine them together. The proposed procedure is essentially distribution-free and computation efficient. The control limit is obtained through the asymptotic distribution of the test statistic under some mild conditions on the dependence structure of stream observations. Our asymptotic results also shed light on quantifying the size of a reference sample required. Both theoretical analysis and numerical results show that the proposed method is able to control the false alarm rate and deliver robust change detection.

scholarly journals Monitoring of mechanical sugarcane harvesting through control charts

2015 ◽  
Vol 35 (6) ◽  
pp. 1079-1092 ◽  
Author(s):  
Murilo A. Voltarelli ◽  
Rouverson P. da Silva ◽  
Cristiano Zerbato ◽  
Carla S. S. Paixão ◽  
Tiago de O. Tavares
Keyword(s):  
Process Control ◽  
Statistical Process Control ◽  
Control Chart ◽  
Control Charts ◽  
Moving Average ◽  
Grinding Wheel ◽  
Individual Values ◽  
Statistical Process ◽  
Minas Gerais State ◽  
Weighted Moving Average

ABSTRACT Statistical process control in mechanized farming is a new way to assess operation quality. In this sense, we aimed to compare three statistical process control tools applied to losses in sugarcane mechanical harvesting to determine the best control chart template for this quality indicator. Losses were daily monitored in farms located within Triângulo Mineiro region, in Minas Gerais state, Brazil. They were carried over a period of 70 days in the 2014 harvest. At the end of the evaluation period, 194 samples were collected in total for each type of loss. The control charts used were individual values chart, moving average and exponentially weighted moving average. The quality indicators assessed during sugarcane harvest were the following loss types: full grinding wheel, stumps, fixed piece, whole cane, chips, loose piece and total losses. The control chart of individual values is the best option for monitoring losses in sugarcane mechanical harvesting, as it is of easier result interpretation, in comparison to the others.

scholarly journals New One-Sided EWMA t Charts without and with Variable Sampling Intervals for Monitoring the Process Mean

2020 ◽  
Vol 2020 ◽  
pp. 1-12
Author(s):  
Yan Wang ◽  
Xuelong Hu ◽  
Xiaojian Zhou ◽  
Yulong Qiao ◽  
Shu Wu
Keyword(s):  
Moving Average ◽  
Simulation Method ◽  
Sampling Interval ◽  
Vital Role ◽  
Process Mean ◽  
Statistical Process ◽  
Variable Sampling ◽  
Sampling Intervals ◽  
Process Variance ◽  
Weighted Moving Average

In statistical process control (SPC), t charts play a vital role in the monitoring of the process mean, especially when the process variance is unknown. In this paper, two separate upper-sided and lower-sided exponentially weighted moving average (EWMA) t charts are first proposed and the Monte Carlo simulation method is used to obtain their run length (RL) properties. Compared with the traditional one-sided EWMA t charts and several run rules t charts, the proposed charts are proven to have better performance than these competing charts. In addition, by adding the variable sampling interval (VSI) feature to the proposed charts, the new VSI one-sided EWMA t charts are shown to detect different shift sizes in the process more efficient than the chart without VSI feature. Finally, an example of a milk filling process illustrates the use of the charts.

scholarly journals Nonparametric EWMA-Type Control Charts for Monitoring Industrial Processes: An Overview

2021 ◽  
Vol 6 (3) ◽  
pp. 708-751
Author(s):  
Ioannis S. Triantafyllou ◽  
Mangey Ram
Keyword(s):  
Probability Distribution ◽  
Statistical Process Control ◽  
Crucial Role ◽  
Control Charts ◽  
Moving Average ◽  
Industrial Processes ◽  
Statistical Process ◽  
Practical Applications ◽  
Memory Type ◽  
Weighted Moving Average

In the present paper we provide an up-to-date overview of nonparametric Exponentially Weighted Moving Average (EWMA) control charts. Due to their nonparametric nature, such memory-type schemes are proved to be very useful for monitoring industrial processes, where the output cannot match to a particular probability distribution. Several fundamental contributions on the topic are mentioned, while recent advances are also presented in some detail. In addition, some practical applications of the nonparametric EWMA-type control charts are highlighted, in order to emphasize their crucial role in the contemporary online statistical process control.

Opening the Black Box: Low-Dimensional Dynamics in High-Dimensional Recurrent Neural Networks

2013 ◽  
Vol 25 (3) ◽  
pp. 626-649 ◽  
Author(s):  
David Sussillo ◽  
Omri Barak
Keyword(s):  
Neural Networks ◽  
Fixed Points ◽  
Recurrent Neural Networks ◽  
Moving Average ◽  
Optimization Technique ◽  
Sine Wave ◽  
High Dimensional ◽  
Flip Flop ◽  
Low Dimensional ◽  
Sine Wave Generator

Recurrent neural networks (RNNs) are useful tools for learning nonlinear relationships between time-varying inputs and outputs with complex temporal dependencies. Recently developed algorithms have been successful at training RNNs to perform a wide variety of tasks, but the resulting networks have been treated as black boxes: their mechanism of operation remains unknown. Here we explore the hypothesis that fixed points, both stable and unstable, and the linearized dynamics around them, can reveal crucial aspects of how RNNs implement their computations. Further, we explore the utility of linearization in areas of phase space that are not true fixed points but merely points of very slow movement. We present a simple optimization technique that is applied to trained RNNs to find the fixed and slow points of their dynamics. Linearization around these slow regions can be used to explore, or reverse-engineer, the behavior of the RNN. We describe the technique, illustrate it using simple examples, and finally showcase it on three high-dimensional RNN examples: a 3-bit flip-flop device, an input-dependent sine wave generator, and a two-point moving average. In all cases, the mechanisms of trained networks could be inferred from the sets of fixed and slow points and the linearized dynamics around them.

scholarly journals Improved Hybrid Exponentially Weighted Moving Average Control Chart using Auxiliary Information

2020 ◽  
Author(s):  
Sadia Tariq ◽  
Muhammad Noor-ul-Amin ◽  
Muhammad Hanif ◽  
Chi-Hyuck Jun 
Keyword(s):  
Control Chart ◽  
Control Charts ◽  
Moving Average ◽  
Real Life ◽  
Auxiliary Information ◽  
Auxiliary Variable ◽  
Exponentially Weighted Moving Average ◽  
Statistical Process ◽  
Weighted Moving Average ◽  
Exponentially Weighted

Statistical process control is an important tool for maintaining the quality of a production process. Several control charts are available to monitor changes in process parameters. In this study, a control chart for the process mean is proposed. For this purpose, an auxiliary variable is used in the form of a regression estimator under the configuration of the hybrid exponentially weighted moving average (HEWMA) control chart. The proposed chart is evaluated by conducting a simulation study. The results showed that the proposed chart is sensitive with respect to the HEWMA chart. A real-life application is also presented to demonstrate the performance of the proposed control chart.

scholarly journals GRAFIK PENGENDALI MIXED EXPONENTIALLY WEIGHTED MOVING AVERAGE – CUMULATIVE SUM (MEC) DALAM ANALISIS PENGAWASAN PROSES PRODUKSI (Studi Kasus : Wingko Babat Cap “Moel”)

2021 ◽  
Vol 10 (1) ◽  
pp. 114-124
Author(s):  
Aulia Resti ◽  
Tatik Widiharih ◽  
Rukun Santoso
Keyword(s):  
Control Chart ◽  
Control Charts ◽  
Average Run Length ◽  
Moving Average ◽  
Exponentially Weighted Moving Average ◽  
Cumulative Sum ◽  
Statistical Process ◽  
Cusum Charts ◽  
Weighted Moving Average ◽  
Exponentially Weighted

Quality control is an important role in industry for maintain quality stability.  Statistical process control can quickly investigate the occurrence of unforeseen causes or process shifts using control charts. Mixed Exponentially Weighted Moving Average - Cumulative Sum (MEC) control chart is a tool used to monitor and evaluate whether the production process is in control or not. The MEC control chart method is a combination of the Exponentially Weighted Moving Average (EWMA) and Cumulative Sum (CUSUM) charts. Combining the two charts aims to increase the sensitivity of the control chart in detecting out of control. To compare the sensitivity level of the EWMA, CUSUM, and MEC methods, the Average Run Length (ARL) was used. From the comparison of ARL values, the MEC chart is the most sensitive control chart in detecting out of control compared to EWMA and CUSUM charts for small shifts. Keywords: Grafik Pengendali, Exponentially Weighted Moving Average, Cumulative Sum, Mixed EWMA-CUSUM, Average Run Lenght, EWMA, CUSUM, MEC, ARL

Time Series Analysis and Statistical Process Control for Anomaly Detection: An Application to the Situation of PM 10 in Northern Thailand

2017 ◽  
Vol 866 ◽  
pp. 379-382
Author(s):  
Unchalee Tonggumnead ◽  
Kittipong Klinjan
Keyword(s):  
Process Control ◽  
Statistical Process Control ◽  
Moving Average ◽  
Correlated Data ◽  
Cusum Chart ◽  
Statistical Process ◽  
Shewhart Chart ◽  
Ewma Chart ◽  
Pm 10 ◽  
Weighted Moving Average

The monitoring of processes is a vital mechanism for ensuring that such processes remain safe and under control. The present research aims to solve problems associated with correlated data by applying the Box-Jenkins method integrated with statistical process control (SPC) tools, namely the Shewhart chart, the moving average chart, the cumulative sum (CUSUM) chart, and the exponentially weighted moving-average (EWMA) chart. The efficiency of the four SPC tools was also compared in terms of the false alarm rate (FAR) and the missed detection rate (MDR). The findings indicated that the EWMA chart was the most effective in detecting anomaly, the Shewhart chart and the moving average chart produced high MDR, and the CUSUM chart suffered the highest FAR.

Optimization Design of 2-EWMA Control Chart Based on Random Process Shift

2013 ◽  
Vol 465-466 ◽  
pp. 1185-1190
Author(s):  
Mohammad Shamsuzzaman
Keyword(s):  
Random Process ◽  
Control Charts ◽  
Optimization Design ◽  
Moving Average ◽  
Rayleigh Distribution ◽  
Statistical Process ◽  
Design Algorithm ◽  
Ewma Chart ◽  
Ewma Control Chart ◽  
Weighted Moving Average

The exponentially weighted moving average (EWMA) control charts are widely used for detecting process shifts of small and moderate sizes in Statistical Process Control (SPC).This article presents an algorithm for the optimization design of a multi-EWMA scheme comprising two EWMA control charts (known as 2-EWMA chart) considering random process shifts in mean. The random process shifts in mean is characterized by a Rayleigh distribution. The design algorithm optimizes the charting parameters of the 2-EWMA chart based on loss function. Comparative study shows that the optimal 2-EWMA chart outperforms the original 2-EWMA chart, as well as the original EWMA chart. In general, this article will help to enhance the detection effectiveness of the 2-EWMA chart, and facilitate its applications in SPC.

Enhanced nonparametric control charts under simple and ranked set sampling schemes

2020 ◽  
Vol 42 (14) ◽  
pp. 2744-2759 ◽  
Author(s):  
Zameer Abbas ◽  
Hafiz Zafar Nazir ◽  
Muhammad Abid ◽  
Noureen Akhtar ◽  
Muhammad Riaz
Keyword(s):  
Statistical Process Control ◽  
Control Charts ◽  
Moving Average ◽  
Exponentially Weighted Moving Average ◽  
Statistical Process ◽  
Sampling Schemes ◽  
Signed Rank ◽  
Process Location ◽  
Weighted Moving Average ◽  
Exponentially Weighted

Investigation and removal of unnatural variation in the processes of manufacturing, production and services require application of statistical process control. Control charts are the most famous and commonly used statistical process control tools to trace changes in the manufacturing and nonmanufacturing processes parameter(s). The nonparametric control charts become necessary when the distribution of underlying process is unknown or questionable. The nonparametric charts are robust alternative along with holding property of quick shift detection ability in process parameter(s). In this article, we have proposed nonparametric double exponentially weighted moving average chart based on Wilcoxon signed rank test under simple and ranked set sampling schemes for efficient monitoring of the process location. The proposed control charts are compared with classical exponentially weighted moving average, double exponentially weighted moving average, nonparametric exponentially weighted moving average sign, nonparametric exponentially weighted moving average signed rank, nonparametric cumulative sum signed rank charts using average run length and some other characteristics of run length distribution as performance measures. Comparison reveals that the proposed control charts performs better to detect all kinds of shifts in the process location than existing counterparts. A real-life application related to manufacturing process (the variable of interest is the diameter of piston ring) is also provided for the practical implementation of the proposed chart.

Sign in / Sign up

Export Citation Format

Share Document