Robust Monitoring of an Industrial IT System in the Presence of Structural Change

Padonou, Espéran; Roustant, Olivier; Lutz, Michel

doi:10.1002/qre.1651

Cited by 5 publications

(5 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…

{\overset{true^}{x}}_{k} = λ_{1} x_{k} + ((), 1 - λ_{1}) ((), {\overset{true^}{x}}_{k - 1} + F_{k - 1})

F_{k} = λ_{2} ((), {\overset{true^}{x}}_{k} - {\overset{true^}{x}}_{k - 1}) + ((), 1 - λ_{2}) F_{k - 1}

where, F k is a variable that estimates the trend and λ 1 and λ 2 are parameters that exponentially weight measurements and the trend estimates, respectively. The filter is widely used for forecasting, assuming that the mean will maintain the same behavior during the prediction horizon

{\overset{true^}{x}}_{k + t | k} = {\overset{true^}{x}}_{k} + t F_{k}

…”

Section: Methodsmentioning

confidence: 99%

“…In their work, the wavelet transform is used to remove outliers, followed by the application of the KF to reconcile measurements with model information. Off‐line denoising is also a very important field, used, for example, in image processing, where the median filter, the hybrid median filter, and the wavelet transform are used extensively . Another category of digital filters is the class of adaptive methods .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Comparison Study of Single‐Scale and Multiscale Approaches for Data‐Driven and Model‐Based Online Denoising

Rendall

Reis

2014

Quality & Reliability Eng

View full text Add to dashboard Cite

Signal denoising is a pervasive operation in most online applications, such as engineering process control and online optimization, strongly affecting the outcome of these higher‐level tasks and impacting the overall variability exhibited by processes and products. Therefore, it plays a fundamental role in improving process capability, which is, however, often overlooked. In this work, we compare the performance of different types of currently available online denoising filters using a variety of test signals that represent the diversity of situations likely to be found in practice, properly corrupted with additive noise of varying magnitudes. Both single‐scale/multiscale, data‐driven/model‐based and time domain/frequency domain, online filtering approaches were contemplated, in what is, to the best of the authors knowledge, the more extensive comparison study conducted on online denoising (or filtering) methodologies. A new class of multiscale denoising algorithms is also considered in this study, based on the online wavelet multiresolution decomposition. In this context, we propose and test a new formulation, called the online multiscale hybrid Kalman filter. After proper tuning, the methods are tested and their performances compared. As a result of the comparison study, clear guidelines are provided for practitioners on the use of online denoising methodologies, which allow for a better management of the impact of the propagation of unstructured components of variability in the final outcome of the processes. Copyright © 2014 John Wiley & Sons, Ltd.

show abstract

“…

{\overset{true^}{x}}_{k} = λ_{1} x_{k} + ((), 1 - λ_{1}) ((), {\overset{true^}{x}}_{k - 1} + F_{k - 1})

F_{k} = λ_{2} ((), {\overset{true^}{x}}_{k} - {\overset{true^}{x}}_{k - 1}) + ((), 1 - λ_{2}) F_{k - 1}

{\overset{true^}{x}}_{k + t | k} = {\overset{true^}{x}}_{k} + t F_{k}

…”

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Comparison Study of Single‐Scale and Multiscale Approaches for Data‐Driven and Model‐Based Online Denoising

Rendall

Reis

2014

Quality & Reliability Eng

View full text Add to dashboard Cite

show abstract

“…Quantitative IT capacity studies are more and more grounded on data analysis methods (allspaw, 2008;Bartolini et al, 2011;Domanski, 1999;Gunther, 2007;Padonou, Roustant, & Lutz, 2015). Indeed, data-driven approaches are particularly well-adapted to manage complex systems where planning time is limited (ahituv & Igbaria, 1988;Gunther, 2007).…”

Section: The Diversity Of Data-driven It Capacity Planning Modelsmentioning

confidence: 99%

“…In this paper, we will restrict the scientific formalisation and contribution to the decision situation 'Definition of a change scenario for the IT architecture and infrastructure' , by showing how a quantitative decision-aid can be provided to size the CPU of the future MeS servers. however, let us mention that numerical solutions were also proposed for 'Identification of any risk of IT capacity failure': an interested reader may refer to Padonou et al (2015).…”

Section: Representation Of the Problem Situation [P]mentioning

confidence: 99%

Data-driven decision-making for IT capacity: beyond statistical analyses

Lutz

Boucher

2016

Journal of Decision Systems

View full text Add to dashboard Cite

International audienceThis paper reports a research work piece developed in collaboration with the semiconductor wafer production company: STMicroelectronics. This collaborative research programme aimed at implementing statistical methods, so as to improve business decisions focusing on capacity planning for information technology. The current paper presents the specification, and experimentation of a method dedicated to managing the need to integrate rigorously complex contextualisation factors, when developing statistical-based Decision Support Systems (DSS). The key challenge is to increase the end-user acceptance and success rate for DSS developments. To ensure the successful integration of DSS within the user environment, the paper formalises a so-called ‘Contextualisation process’, integrated within a larger decision-aid development framework. This Contextualisation process is specified with three methodological components, respectively ‘Qualitative Contextualisation’, ‘Statistical Modelling and Formalisation’ and ‘User Integration’. This approach is applied by STMicroelectronics, to a case study focusing on a project of change management for the infrastructure of the Information System. Based on the demonstration of the case study results, the added-value of the Contextualisation process are discussed and further perspectives for applied research in DSS are drawn

show abstract

“…1) One type of methods is based on traditional time sequence forecasting model, such as: auto-regressive integrated moving average model [5,6] , index smoothing forecasting method [7][8][9] and so on. In particular, Fang et al analyzed real-time plant data via a principal component analysis method and proposed a distributed and hierarchical monitoring framework for early warnings [10] .…”

Section: Introductionmentioning

confidence: 99%

Alarm state forecasting from similar trends in historical time sequences based on Bayesian estimators

Kang

et al. 2023

Second International Conference on Advanced Manufacturing Technology and Manufacturing Systems (ICAMTMS 2023)

View full text Add to dashboard Cite

Predicting the upcoming alarm state helps industrial plant operators take timely actions to avoid the negative effects of abnormal conditions. This paper proposes a new data-driven method for trend forecasting in a process variable. The proposed method has two important features: (i) alarm states are predicted in real-time whenever a new trend occurs in the current time sequence, which helps industrial operators take timely actions at the early stage of abnormal conditions; (ii) alarm states are predicted with quantitative probabilities and credibility intervals. By contrast, existing early warning methods are either suitable for specific scenarios where a short prediction step length is needed or cannot provide a quantitative measurement of prediction results. The proposed method firstly captures qualitative trends in the current and historical time sequences and clusters these trends into several categories. Secondly, current and historical trend sequences are obtained according to the cluster results. Finally, recent trends in the current trend sequence are chosen, and alarm states are predicted based on the historical trend sequence having the same sub-sequence. A Bayesian estimator is exploited to calculate the posterior probabilities and their credibility intervals for predicted alarm states. The optimal predicted alarm state is the one with the largest lower bound of posterior probability as a quantitative measurement of prediction accuracy. The effectiveness of the proposed method is illustrated via numerical examples based on a well-accepted benchmark of the Tennessee Eastman process.

show abstract

Robust Monitoring of an Industrial IT System in the Presence of Structural Change

Cited by 5 publications

References 48 publications

A Comparison Study of Single‐Scale and Multiscale Approaches for Data‐Driven and Model‐Based Online Denoising

A Comparison Study of Single‐Scale and Multiscale Approaches for Data‐Driven and Model‐Based Online Denoising

Data-driven decision-making for IT capacity: beyond statistical analyses

Alarm state forecasting from similar trends in historical time sequences based on Bayesian estimators

Contact Info

Product

Resources

About