Abnormal situation management: Challenges and opportunities in the big data era

Shu, Yidan; Liang, Ming; Fang, Cheng; Zhang, Zhanpeng; Zhao, Jun

doi:10.1016/j.compchemeng.2016.04.011

Cited by 79 publications

(40 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classification is the most common approach at the predictive analytics level and has been widely applied in manufacturing to support production planning and control (Chien et al, 2014; and equipment maintenance and diagnosis Shu et al, 2016;. This type of BDA model also plays a key role in logistics/transportation (Li et al, 2014;Yu and AbdelAty, 2014;Zangenehpour et al, 2015) and procurement research (Ling Ho and Wen Shih, 2014;Mori et al, 2012) but apparently, current studies in those areas have not fully exploited the advantages of classification.…”

Section: Resultsmentioning

confidence: 99%

“…Product R&D Bae and Kim, 2011;Do, 2014;Lei and Moon, 2015;Opresnik and Taisch, 2015;Zhang et al, 2017 Production planning & control Chien et al, 2014;Krumeich et al, 2016;Zhang et al, 2017;Shu et al, 2016;Dai et al, 2012;Zhong et al, 2016;Helo and Hao, 2017 Quality management Krumeich et al, 2016;Zhang et al, 2017; Maintenance & diagnosis Shu et al, 2016;Guo et al, 2016;Kumar et al, 2016;Zhang et al, 2017;Wang et al, 2015 Warehousing Storage assignment Chuang et al, 2014;Li, Moghaddam, et al, 2016;Tsai and Huang, 2015;Chiang et al, 2011;Chiang et al, 2014 Order picking Ballestín et al, 2013;Chuang et al, 2014; Inventory control Alyahya et al, 2016;Hofmann, 2015;Huang and Van Mieghem, 2014;Lee et al, 2016;Stefanovic, 2015 Logistics and transportation Cui et al, 2016;Li et al, 2015;Shi and Abdel-Aty, 2015;StAubin et al, 2015;Toole et al, 2015;Xia et al, 2016;Yu and Abdel-Aty, 2014;Zangenehpour et al, 2015;…”

Section: Manufacturingmentioning

confidence: 99%

See 1 more Smart Citation

Big data analytics in supply chain management: A state-of-the-art literature review

Nguyen

Zhou

Spiegler

et al. 2018

Computers & Operations Research

406

238

View full text Add to dashboard Cite

Abstract:The rapidly growing interest from both academics and practitioners in the application of big data analytics (BDA) in supply chain management (SCM) has urged the need for review of up-to-date research development in order to develop a new agenda. This review responds to the call by proposing a novel classification framework that provides a full picture of current literature on where and how BDA has been applied within the SCM context. The classification framework is structurally based on the content analysis method of Mayring (2008) techniques are employed to develop these models? The discussion tackling these four questions reveals a number of research gaps, which leads to future research directions.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Manufacturingmentioning

confidence: 99%

Big data analytics in supply chain management: A state-of-the-art literature review

Nguyen

Zhou

Spiegler

et al. 2018

Computers & Operations Research

406

238

View full text Add to dashboard Cite

show abstract

“…This process was repeated until there was no significant performance improvement. Overall performance evaluation for both training and test phases were measured by the functions sum of squared error (SOS) and cross entropy (CE), described in Equations (5) and (6). The percentage of correct classifications for each class is presented as a confusion matrix:…”

Section: Methodsmentioning

confidence: 99%

“…Venkatasubramanian, Rengaswamy, and Kavuri [3][4][5] summarize and review the main characteristics of the techniques described in the literature. Nonetheless, the size and complexity of chemical process industries and the increasing amount of data available in this digital era endorse 2 of 14 the use of data-based FDD techniques, such as neural network algorithms and multivariate statistical methods [6]. Ragabab and co-workers, Casanova-Peláez and co-workers, and Leiviskä [7][8][9] are some examples of data-driven FDD applications in chemical processes applications.…”

Section: Introductionmentioning

confidence: 99%

Data Augmentation Applied to Machine Learning-Based Monitoring of a Pulp and Paper Process

et al. 2019

View full text Add to dashboard Cite

Industrial archived process data represent a convenient source of information for data-driven models, such as artificial neural network (ANN), that can be used for safety and efficiency improvement like early or even predictive fault detection and diagnosis (FDD). Nonetheless, most of the data used for model generation are representative of the process nominal states and therefore are not enough for classification problems intended to determine abnormal process conditions. This work proposes the use of techniques to augment the original real data standards, dismissing the need for experiments that could jeopardize process safety. It uses the Monte Carlo technique to artificially increase the number of model inputs coupled to the nearest neighbor search (NNS) by geometric distances to consistently classify the generated patterns in normal or faulty statuses. Finally, a radial basis function neural network is trained with the augmented data. The methodology was validated by a study case in which 3381 pulp and paper industrial data points were expanded to monitor the formation of particles in a recovery boiler. Only 5.8% of the original process data were examples of faulty conditions, but the new expanded and balanced data collection leveraged the classification performance of the neural network, allowing its future use for monitoring purpose.

show abstract

“…Techniques such as real time fault detection and diagnosis (FDD) play a key role in avoiding major accidents, enhancing equipment reliability, and plant profitability. Due to proliferation of historical data, the process history based methods have gained grounds over the model based methods [4].…”

Section: Introductionmentioning

confidence: 99%

Abnormal Situation Management: Detection of Fouling and Longitudinal Bypass in Heat Exchangers

Swan¹,

Fernandes²,

Khade³

et al. 2018

IJCEA

View full text Add to dashboard Cite

Improved condition monitoring of production equipment enhances process safety and the ability to perform preventive maintenance, thus reduces downtime and associated costs. Condition monitoring or early detection is an integrated part of Abnormal Situation Management (ASM) and is becoming more and more important to chemical process industries due to safety, environmental, and economic reasons. Despite these developments, ASM is generally poorly practiced in the process industries. Much of the actual practice places the onus on plant operators to respond effectively to abnormal conditions. The objective of this article is to plot the data in real time for the detection of fouling in air-cooled exchangers and leaks due to longitudinal bypass in shell and tube heat exchangers. This data is obtained from the Distributive Control System (DCS) historian in the case of shell and tube heat exchanger and experimental methods for air cooled heat exchangers respectively. In this research apparent fouling in one of the shell and tube heat exchanger was actually longitudinal baffle bypass leak. The methodology used in this study is non-intrusive condition monitoring of the heat exchangers of a petrochemical plant. For the air cooled heat exchangers, the data is obtained by using a hot wire anemometer to measure the air flow and thermocouple wire to measure the temperature of the air for both inlet and outlet conditions respectively. The data got is then analyzed by using heat transfer equations and ASPEN exchanger design program. In the case of shell and tube heat exchangers the flow rate and temperature data of both the shell side and tube side fluid is got from the DCS historian respectively. Then using the heat transfer equations for a shell and tube heat exchanger the data is plotted and analyzed. The results obtained indicate that fouling in the case of air cooled heat exchangers takes place on the air side and is higher in carbon steel tubes than in admiralty. In the case of shell and tube heat exchangers for the BFU type; apparent fouling is actually the longitudinal bypass leak which is severe. For the BEU type fouling, it is due to gum formation and is verified from the data. The schemes for early detection of the above are expected to be a useful for the development of a diagnostic tool for operation alert for ASM and preventive maintenance respectively.

show abstract

Abnormal situation management: Challenges and opportunities in the big data era

Cited by 79 publications

References 72 publications

Big data analytics in supply chain management: A state-of-the-art literature review

Big data analytics in supply chain management: A state-of-the-art literature review

Data Augmentation Applied to Machine Learning-Based Monitoring of a Pulp and Paper Process

Abnormal Situation Management: Detection of Fouling and Longitudinal Bypass in Heat Exchangers

Contact Info

Product

Resources

About