Topical Collection "Process Data Analytics"

Editors

Collection Editor
Dr. Leo H. Chiang

Dow Chemical Company, USA
Website | E-Mail
Interests: process data analytics; machine learning; big data; visualization; process monitoring; and Industry 4.0
Collection Editor
Prof. Richard D. Braatz

Massachusetts Institute of Technology, USA
Website | E-Mail

Topical Collection Information

Dear Colleagues,

Data analytics is a term used to describe a set of computational methods for the analysis of data to abstract knowledge and affect decision making. Typical information of interest includes (1) uncovering unknown patterns or correlations within the data; (2) constructing predictions of some variables as functions of other variables; (3) identifying data points that are atypical of the overall dataset; and (4) classifying different groups of outliers. The development of data analytics methods has seen rapid growth in the last decade, primarily by the machine learning and related communities that formulate answers to specific questions in terms of optimization problems.

This Special Issue concerns process data analytics, which refers to data analytics methods that are suitable for the types of data and problems that arise in manufacturing processes. The quantity of process data that has become available and stored in historical databases for manufacturing processes has grown by orders of magnitude, but the abstraction of the most value from this data has been elusive. The commonly used tools used in industrial practice have significant limitations in utility and performance, to such an extent that most data stored in historical databases are not analyzed at all rather than being analyzed poorly. Tools from machine learning and related communities typically require significant modifications to be effective for process data, and the structure of the available prior mechanistic information and other domain knowledge on processes and the types of questions that arise in manufacturing processes have a specificity that need to be taken into account to be able to develop the most effective data analytics methods.

This Special Issue, ”Process Data Analytics”, aims to bring together recent advances, and invites all original contributions, fundamental and applied, which can add to our understanding of the field. Topics may include, but are not limited to:

  • Process data analytics methods
  • Machine learning methods adapted for application to manufacturing processes
  • Methods for better handling of missing data
  • Fault detection and diagnosis
  • Adaptive process monitoring
  • Industrial case studies
  • Applications to Big Data problems in manufacturing
  • Hybrid data analytics methods
  • Prognostic systems

Dr. Leo H. Chiang
Prof. Richard D. Braatz
Collection Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the collection website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Processes is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1100 CHF (Swiss Francs). Please note that for papers submitted after 30 June 2019 an APC of 1200 CHF applies. Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Data analytics
  • Process data analytics
  • Big data
  • Big data analytics
  • Machine learning
  • Diagnostic systems
  • Prognostics
  • Process monitoring
  • Process health monitoring
  • Fault detection and diagnosis

Published Papers (12 papers)

2018

Jump to: 2017

Open AccessArticle Centrifugal Pump Monitoring and Determination of Pump Characteristic Curves Using Experimental and Analytical Solutions
Processes 2018, 6(2), 18; https://doi.org/10.3390/pr6020018
Received: 12 December 2017 / Revised: 7 February 2018 / Accepted: 9 February 2018 / Published: 13 February 2018
PDF Full-text (3566 KB) | HTML Full-text | XML Full-text
Abstract
Centrifugal pumps are widely used in the industry, especially in the oil and gas sector for fluids transport. Classically, these are designed to transfer single phase fluids (e.g., water) at high flow rates and relatively low pressures when compared with other pump types. [...] Read more.
Centrifugal pumps are widely used in the industry, especially in the oil and gas sector for fluids transport. Classically, these are designed to transfer single phase fluids (e.g., water) at high flow rates and relatively low pressures when compared with other pump types. As part of their constructive feature, centrifugal pumps rely on seals to prevent air entrapment into the rotor during its normal operation. Although this is a constructive feature, water should pass through the pump inlet even when the inlet manifold is damaged. Modern pumps are integrated in pumping units which consist of a drive (normally electric motor), a transmission (when needed), an electronic package (for monitoring and control), and the pump itself. The unit also has intake and outlet manifolds equipped with valves. Modern systems also include electronic components to measure and monitor pump working parameters such as pressure, temperature, etc. Equipment monitoring devices (vibration sensors, microphones) are installed on modern pumping units to help users evaluate the state of the machinery and detect deviations from the normal working condition. This paper addresses the influence of air-water two-phase mixture on the characteristic curve of a centrifugal pump; pump vibration in operation at various flow rates under these conditions; the possibilities of using the results of experimental investigations in the numerical simulations for design and training purposes, and the possibility of using vibration and sound analysis to detect changes in the equipment working condition. Conclusions show that vibration analysis provides accurate information about the pump’s functional state and the pumping process. Moreover, the acoustic emission also enables the evaluation of the pump status, but needs further improvements to better capture and isolate the usable sounds from the environment. Full article
Figures

Figure 1

Open AccessFeature PaperArticle Predicting the Operating States of Grinding Circuits by Use of Recurrence Texture Analysis of Time Series Data
Processes 2018, 6(2), 17; https://doi.org/10.3390/pr6020017
Received: 28 December 2017 / Revised: 24 January 2018 / Accepted: 29 January 2018 / Published: 11 February 2018
Cited by 1 | PDF Full-text (7088 KB) | HTML Full-text | XML Full-text
Abstract
Grinding circuits typically contribute disproportionately to the overall cost of ore beneficiation and their optimal operation is therefore of critical importance in the cost-effective operation of mineral processing plants. This can be challenging, as these circuits can also exhibit complex, nonlinear behavior that [...] Read more.
Grinding circuits typically contribute disproportionately to the overall cost of ore beneficiation and their optimal operation is therefore of critical importance in the cost-effective operation of mineral processing plants. This can be challenging, as these circuits can also exhibit complex, nonlinear behavior that can be difficult to model. In this paper, it is shown that key time series variables of grinding circuits can be recast into sets of descriptor variables that can be used in advanced modelling and control of the mill. Two real-world case studies are considered. In the first, it is shown that the controller states of an autogenous mill can be identified from the load measurements of the mill by using a support vector machine and the abovementioned descriptor variables as predictors. In the second case study, it is shown that power and temperature measurements in a horizontally stirred mill can be used for online estimation of the particle size of the mill product. Full article
Figures

Figure 1

Open AccessArticle A Throughput Management System for Semiconductor Wafer Fabrication Facilities: Design, Systems and Implementation
Processes 2018, 6(2), 16; https://doi.org/10.3390/pr6020016
Received: 16 December 2017 / Revised: 2 February 2018 / Accepted: 9 February 2018 / Published: 11 February 2018
PDF Full-text (1759 KB) | HTML Full-text | XML Full-text
Abstract
Equipment throughput is one of the most critical parameters for production planning and scheduling, which is often derived by optimization techniques to achieve business goals. However, in semiconductor manufacturing, up-to-date and reliable equipment throughput is not easy to estimate and maintain because of [...] Read more.
Equipment throughput is one of the most critical parameters for production planning and scheduling, which is often derived by optimization techniques to achieve business goals. However, in semiconductor manufacturing, up-to-date and reliable equipment throughput is not easy to estimate and maintain because of the high complexity and extreme amount of data in the production systems. This article concerns the development and implementation of a throughput management system tailored for a semiconductor wafer fabrication plant (Fab). A brief overview of the semiconductor manufacturing and an introduction of the case Fab are presented first. Then, we focus on the system architecture and some concepts of crucial modules. This study also describes the project timescales and difficulties and discusses both tangible and intangible benefits from this project. Full article
Figures

Figure 1

2017

Jump to: 2018

Open AccessArticle RadViz Deluxe: An Attribute-Aware Display for Multivariate Data
Processes 2017, 5(4), 75; https://doi.org/10.3390/pr5040075
Received: 12 October 2017 / Revised: 8 November 2017 / Accepted: 17 November 2017 / Published: 22 November 2017
Cited by 1 | PDF Full-text (6211 KB) | HTML Full-text | XML Full-text
Abstract
Modern data, such as occurring in chemical engineering, typically entail large collections of samples with numerous dimensional components (or attributes). Visualizing the samples in relation of these components can bring valuable insight. For example, one may be able to see how a [...] Read more.
Modern data, such as occurring in chemical engineering, typically entail large collections of samples with numerous dimensional components (or attributes). Visualizing the samples in relation of these components can bring valuable insight. For example, one may be able to see how a certain chemical property is expressed in the samples taken. This could reveal if there are clusters and outliers that have specific distinguishing properties. Current multivariate visualization methods lack the ability to reveal these types of information at a sufficient degree of fidelity since they are not optimized to simultaneously present the relations of the samples as well as the relations of the samples to their attributes. We propose a display that is designed to reveal these multiple relations. Our scheme is based on the concept of RadViz, but enhances the layout with three stages of iterative refinement. These refinements reduce the layout error in terms of three essential relationships—sample to sample, attribute to attribute, and sample to attribute. We demonstrate the effectiveness of our method via various real-world domain examples in the domain of chemical process engineering. In addition, we also formally derive the equivalence of RadViz to a popular multivariate interpolation method called generalized barycentric coordinates. Full article
Figures

Figure 1

Open AccessArticle How to Generate Economic and Sustainability Reports from Big Data? Qualifications of Process Industry
Processes 2017, 5(4), 64; https://doi.org/10.3390/pr5040064
Received: 2 October 2017 / Revised: 23 October 2017 / Accepted: 24 October 2017 / Published: 1 November 2017
Cited by 2 | PDF Full-text (1582 KB) | HTML Full-text | XML Full-text
Abstract
Big Data may introduce new opportunities, and for this reason it has become a mantra among most industries. This paper focuses on examining how to develop cost and sustainable reporting by utilizing Big Data that covers economic values, production volumes, and emission information. [...] Read more.
Big Data may introduce new opportunities, and for this reason it has become a mantra among most industries. This paper focuses on examining how to develop cost and sustainable reporting by utilizing Big Data that covers economic values, production volumes, and emission information. We assume strongly that this use supports cleaner production, while at the same time offers more information for revenue and profitability development. We argue that Big Data brings company-wide business benefits if data queries and interfaces are built to be interactive, intuitive, and user-friendly. The amount of information related to operations, costs, emissions, and the supply chain would increase enormously if Big Data was used in various manufacturing industries. It is essential to expose the relevant correlations between different attributes and data fields. Proper algorithm design and programming are key to making the most of Big Data. This paper introduces ideas on how to refine raw data into valuable information, which can serve many types of end users, decision makers, and even external auditors. Concrete examples are given through an industrial paper mill case, which covers environmental aspects, cost-efficiency management, and process design. Full article
Figures

Figure 1

Open AccessArticle A Long-Short Term Memory Recurrent Neural Network Based Reinforcement Learning Controller for Office Heating Ventilation and Air Conditioning Systems
Processes 2017, 5(3), 46; https://doi.org/10.3390/pr5030046
Received: 13 June 2017 / Revised: 10 August 2017 / Accepted: 10 August 2017 / Published: 18 August 2017
Cited by 9 | PDF Full-text (1647 KB) | HTML Full-text | XML Full-text
Abstract
Energy optimization in buildings by controlling the Heating Ventilation and Air Conditioning (HVAC) system is being researched extensively. In this paper, a model-free actor-critic Reinforcement Learning (RL) controller is designed using a variant of artificial recurrent neural networks called Long-Short-Term Memory (LSTM) networks. [...] Read more.
Energy optimization in buildings by controlling the Heating Ventilation and Air Conditioning (HVAC) system is being researched extensively. In this paper, a model-free actor-critic Reinforcement Learning (RL) controller is designed using a variant of artificial recurrent neural networks called Long-Short-Term Memory (LSTM) networks. Optimization of thermal comfort alongside energy consumption is the goal in tuning this RL controller. The test platform, our office space, is designed using SketchUp. Using OpenStudio, the HVAC system is installed in the office. The control schemes (ideal thermal comfort, a traditional control and the RL control) are implemented in MATLAB. Using the Building Control Virtual Test Bed (BCVTB), the control of the thermostat schedule during each sample time is implemented for the office in EnergyPlus alongside local weather data. Results from training and validation indicate that the RL controller improves thermal comfort by an average of 15% and energy efficiency by an average of 2.5% as compared to other strategies mentioned. Full article
Figures

Figure 1

Open AccessArticle Data Visualization and Visualization-Based Fault Detection for Chemical Processes
Processes 2017, 5(3), 45; https://doi.org/10.3390/pr5030045
Received: 2 June 2017 / Revised: 18 July 2017 / Accepted: 31 July 2017 / Published: 14 August 2017
PDF Full-text (593 KB) | HTML Full-text | XML Full-text
Abstract
Over the years, there has been a consistent increase in the amount of data collected by systems and processes in many different industries and fields. Simultaneously, there is a growing push towards revealing and exploiting of the information contained therein. The chemical processes [...] Read more.
Over the years, there has been a consistent increase in the amount of data collected by systems and processes in many different industries and fields. Simultaneously, there is a growing push towards revealing and exploiting of the information contained therein. The chemical processes industry is one such field, with high volume and high-dimensional time series data. In this paper, we present a unified overview of the application of recently-developed data visualization concepts to fault detection in the chemical industry. We consider three common types of processes and compare visualization-based fault detection performance to methods used currently. Full article
Figures

Figure 1

Open AccessFeature PaperReview Design of Experiments for Control-Relevant Multivariable Model Identification: An Overview of Some Basic Recent Developments
Processes 2017, 5(3), 42; https://doi.org/10.3390/pr5030042
Received: 22 May 2017 / Revised: 26 July 2017 / Accepted: 29 July 2017 / Published: 3 August 2017
Cited by 2 | PDF Full-text (5016 KB) | HTML Full-text | XML Full-text
Abstract
The effectiveness of model-based multivariable controllers depends on the quality of the model used. In addition to satisfying standard accuracy requirements for model structure and parameter estimates, a model to be used in a controller must also satisfy control-relevant requirements, such as integral [...] Read more.
The effectiveness of model-based multivariable controllers depends on the quality of the model used. In addition to satisfying standard accuracy requirements for model structure and parameter estimates, a model to be used in a controller must also satisfy control-relevant requirements, such as integral controllability. Design of experiments (DOE), which produce data from which control-relevant models can be accurately estimated, may differ from standard DOE. The purpose of this paper is to emphasize this basic principle and to summarize some fundamental results obtained in recent years for DOE in two important cases: Accurate estimation of the order of a multivariable model and efficient identification of a model that satisfies integral controllability; both important for the design of robust model-based controllers. For both cases, we provide an overview of recent results that can be easily incorporated by the final user in related DOE. Computer simulations illustrate outcomes to be anticipated. Finally, opportunities for further development are discussed. Full article
Figures

Figure 1

Open AccessFeature PaperArticle Big Data Analytics for Smart Manufacturing: Case Studies in Semiconductor Manufacturing
Processes 2017, 5(3), 39; https://doi.org/10.3390/pr5030039
Received: 6 June 2017 / Revised: 28 June 2017 / Accepted: 4 July 2017 / Published: 12 July 2017
Cited by 14 | PDF Full-text (4390 KB) | HTML Full-text | XML Full-text
Abstract
Smart manufacturing (SM) is a term generally applied to the improvement in manufacturing operations through integration of systems, linking of physical and cyber capabilities, and taking advantage of information including leveraging the big data evolution. SM adoption has been occurring unevenly across industries, [...] Read more.
Smart manufacturing (SM) is a term generally applied to the improvement in manufacturing operations through integration of systems, linking of physical and cyber capabilities, and taking advantage of information including leveraging the big data evolution. SM adoption has been occurring unevenly across industries, thus there is an opportunity to look to other industries to determine solution and roadmap paths for industries such as biochemistry or biology. The big data evolution affords an opportunity for managing significantly larger amounts of information and acting on it with analytics for improved diagnostics and prognostics. The analytics approaches can be defined in terms of dimensions to understand their requirements and capabilities, and to determine technology gaps. The semiconductor manufacturing industry has been taking advantage of the big data and analytics evolution by improving existing capabilities such as fault detection, and supporting new capabilities such as predictive maintenance. For most of these capabilities: (1) data quality is the most important big data factor in delivering high quality solutions; and (2) incorporating subject matter expertise in analytics is often required for realizing effective on-line manufacturing solutions. In the future, an improved big data environment incorporating smart manufacturing concepts such as digital twin will further enable analytics; however, it is anticipated that the need for incorporating subject matter expertise in solution design will remain. Full article
Figures

Figure 1

Open AccessFeature PaperArticle Principal Component Analysis of Process Datasets with Missing Values
Processes 2017, 5(3), 38; https://doi.org/10.3390/pr5030038
Received: 31 May 2017 / Revised: 28 June 2017 / Accepted: 30 June 2017 / Published: 6 July 2017
Cited by 1 | PDF Full-text (924 KB) | HTML Full-text | XML Full-text
Abstract
Datasets with missing values arising from causes such as sensor failure, inconsistent sampling rates, and merging data from different systems are common in the process industry. Methods for handling missing data typically operate during data pre-processing, but can also occur during model building. [...] Read more.
Datasets with missing values arising from causes such as sensor failure, inconsistent sampling rates, and merging data from different systems are common in the process industry. Methods for handling missing data typically operate during data pre-processing, but can also occur during model building. This article considers missing data within the context of principal component analysis (PCA), which is a method originally developed for complete data that has widespread industrial application in multivariate statistical process control. Due to the prevalence of missing data and the success of PCA for handling complete data, several PCA algorithms that can act on incomplete data have been proposed. Here, algorithms for applying PCA to datasets with missing values are reviewed. A case study is presented to demonstrate the performance of the algorithms and suggestions are made with respect to choosing which algorithm is most appropriate for particular settings. An alternating algorithm based on the singular value decomposition achieved the best results in the majority of test cases involving process datasets. Full article
Figures

Figure 1

Open AccessArticle Industrial Process Monitoring in the Big Data/Industry 4.0 Era: from Detection, to Diagnosis, to Prognosis
Processes 2017, 5(3), 35; https://doi.org/10.3390/pr5030035
Received: 1 June 2017 / Revised: 25 June 2017 / Accepted: 27 June 2017 / Published: 30 June 2017
Cited by 20 | PDF Full-text (773 KB) | HTML Full-text | XML Full-text
Abstract
We provide a critical outlook of the evolution of Industrial Process Monitoring (IPM) since its introduction almost 100 years ago. Several evolution trends that have been structuring IPM developments over this extended period of time are briefly referred, with more focus on data-driven [...] Read more.
We provide a critical outlook of the evolution of Industrial Process Monitoring (IPM) since its introduction almost 100 years ago. Several evolution trends that have been structuring IPM developments over this extended period of time are briefly referred, with more focus on data-driven approaches. We also argue that, besides such trends, the research focus has also evolved. The initial period was centred on optimizing IPM detection performance. More recently, root cause analysis and diagnosis gained importance and a variety of approaches were proposed to expand IPM with this new and important monitoring dimension. We believe that, in the future, the emphasis will be to bring yet another dimension to IPM: prognosis. Some perspectives are put forward in this regard, including the strong interplay of the Process and Maintenance departments, hitherto managed as separated silos. Full article
Figures

Figure 1

Open AccessFeature PaperArticle Outlier Detection in Dynamic Systems with Multiple Operating Points and Application to Improve Industrial Flare Monitoring
Processes 2017, 5(2), 28; https://doi.org/10.3390/pr5020028
Received: 25 March 2017 / Revised: 8 May 2017 / Accepted: 24 May 2017 / Published: 31 May 2017
Cited by 2 | PDF Full-text (862 KB) | HTML Full-text | XML Full-text
Abstract
In chemical industries, process operations are usually comprised of several discrete operating regions with distributions that drift over time. These complexities complicate outlier detection in the presence of intrinsic process dynamics. In this article, we consider the problem of detecting univariate outliers in [...] Read more.
In chemical industries, process operations are usually comprised of several discrete operating regions with distributions that drift over time. These complexities complicate outlier detection in the presence of intrinsic process dynamics. In this article, we consider the problem of detecting univariate outliers in dynamic systems with multiple operating points. A novel method combining the time series Kalman filter (TSKF) with the pruned exact linear time (PELT) approach to detect outliers is proposed. The proposed method outperformed benchmark methods in outlier removal performance using simulated data sets of dynamic systems with mean shifts. The method was also able to maintain the integrity of the original data set after performing outlier removal. In addition, the methodology was tested on industrial flaring data to pre-process the flare data for discriminant analysis. The industrial test case shows that performing outlier removal dramatically improves flare monitoring results through Partial Least Squares Discriminant Analysis (PLS-DA), which further confirms the importance of data cleaning in process data analytics. Full article
Figures

Figure 1

Processes EISSN 2227-9717 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top