Hybrid approach to casual analysis on a complex industrial system based on transfer entropy in conjunction with process connectivity information

Landman, Rinat; Jämsä‐Jounela, Sirkka‐Liisa

doi:10.1016/j.conengprac.2016.04.010

Cited by 41 publications

(16 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, undesired conditions in one unit can potentially have plant-wide effects. These conditions deteriorate the product quality, increase operational costs, and can lead to hazardous situations [59]. In these scenarios, effective root cause analysis of these undesired conditions is crucial for restoring the process to its normal operating condition in a timely manner [32].…”

Section: Root Cause Analysismentioning

confidence: 99%

“…Application: Due to their ability to detect causal direction, causal discovery methods are especially useful in manufacturing applications of root cause analysis. Several approaches have focused on combining process topology and connectivity information to improve accuracy and reduce the computational load of root cause analysis in industrial board and board machine case studies [59][60][61][62][63]. Other studies demonstrated the application of causal discovery methods in the detection of disturbance propagation paths in fluid catalytic cracking units [64], mineral concentrator plants [65], and semiconductor production facilities [66].…”

Section: Root Cause Analysismentioning

confidence: 99%

See 1 more Smart Citation

Causal Discovery in Manufacturing: A Structured Literature Review

Vuković

Thalmann

2022

JMMP

View full text Add to dashboard Cite

Industry 4.0 radically alters manufacturing organization and management, fostering collection and analysis of increasing amounts of data. Advanced data analytics, such as machine learning (ML), are essential for implementing Industry 4.0 and obtaining insights regarding production, better decision support, and enhanced manufacturing quality and sustainability. ML outperforms traditional approaches in many cases, but its complexity leads to unclear bases for decisions. Thus, acceptance of ML and, concomitantly, Industry 4.0, is hindered due to increasing requirements of fairness, accountability, and transparency, especially in sensitive-use cases. ML does not augment organizational knowledge, which is highly desired by manufacturing experts. Causal discovery promises a solution by providing insights on causal relationships that go beyond traditional ML’s statistical dependency. Causal discovery has a theoretical background and been successfully applied in medicine, genetics, and ecology. However, in manufacturing, only experimental and scattered applications are known; no comprehensive overview about how causal discovery can be applied in manufacturing is available. This paper investigates the state and development of research on causal discovery in manufacturing by focusing on motivations for application, common application scenarios and approaches, impacts, and implementation challenges. Based on the structured literature review, four core areas are identified, and a research agenda is proposed.

show abstract

Section: Root Cause Analysismentioning

confidence: 99%

Section: Root Cause Analysismentioning

confidence: 99%

Causal Discovery in Manufacturing: A Structured Literature Review

Vuković

Thalmann

2022

JMMP

View full text Add to dashboard Cite

show abstract

“…Second, for all feasible solutions {Z n } N n=1 with N in ( 14), the explanatory ratios {α n } N n=1 are calculated from (15). Fig.…”

Section: Industrial Case Studymentioning

confidence: 99%

“…Chiang et al [14] exploited the data-driven and causal connectivity-based features as well as the propagation path-based feature for diagnosing faults. Landman and Jamsa-Jounela [15], [16] proposed new hybrid approaches for detecting causality based on transfer entropies and nearest neighbors by incorporating process connectivity information. Taktak et al [17] diagnosed abrupt parametric faults of switched systems based on time series data abstraction and a hybrid bond graph model.…”

Section: Introductionmentioning

confidence: 99%

Selection of Root-Cause Process Variables Based on Qualitative Trends in Historical Data Samples

Wang

Chen

2019

IEEE Access

View full text Add to dashboard Cite

Root cause analysis helps industrial plant operators in finding possible root causes of alarms and their associated abnormalities, where one pre-requisite information is the set of root-cause process variables. This paper proposes a method to determine a set of root-cause process variables for a primary process variable based on their qualitative trends. According to requirements on time durations, amplitude changes and correlation coefficients, qualitative trends of process variables are extracted from historical data samples via a dynamic programming approach. A set of root-cause process variables is selected as the one whose qualitative trend combinations are associated with the largest ratio in explaining increasing and decreasing trends of the primary process variable. An industrial case study is provided to illustrate the effectiveness of the proposed method.INDEX TERMS Root cause analysis, qualitative trend analysis, industrial alarm systems.

show abstract

“…Extension of the Granger causality to nonlinear process systems where the nonlinearity can stem forth due to interaction between the input and state variables as well as due to interaction among the state variables is not straight forward. The use of transfer entropy for measuring process connectivity for fault diagnosis including process connectivity has been reported (Landman and Jämsä-Jounela, 2016). These two methods are typically applied when the variables are assumed to be Gaussian (Barnett et al, 2009).…”

Section: Journal Publicationsmentioning

confidence: 99%

Biomimetic and Advanced Control Structure Design with Real Time Optimization

Bankole¹

View full text Add to dashboard Cite

show abstract

Hybrid approach to casual analysis on a complex industrial system based on transfer entropy in conjunction with process connectivity information

Cited by 41 publications

References 24 publications

Causal Discovery in Manufacturing: A Structured Literature Review

Causal Discovery in Manufacturing: A Structured Literature Review

Selection of Root-Cause Process Variables Based on Qualitative Trends in Historical Data Samples

Biomimetic and Advanced Control Structure Design with Real Time Optimization

Contact Info

Product

Resources

About