Fault diagnosis on production systems with support vector machine and decision trees algorithms

Demetgül, Mustafa

doi:10.1007/s00170-012-4639-5

Cited by 46 publications

(33 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SVM uses different kernel functions like radial basis function (RBF) or polynomial kernel to find a hyperplane that best separates data into their classes, and has good classification performance when used with small training sets [23]. Successful areas of application of SVM range from face recognition, recognition of handwritten characters, speech recognition, image retrieval, prediction, etc.…”

Section: Support Vector Machinementioning

confidence: 99%

“…Kumar et al [27] created a MapReduce framework for automatic diagnosis for cloud based manufacturing using SVM as the classification algorithm and validated this with a case study of fault diagnosis using the steel plate manufacturing data available on UCI Machine Learning Repository [28]. Demetgul [23] used SVM to classify 9 fault states in a modular production system (MPS) using data obtained from eight sensors, and experimented with 4 different kernel functions namely RBF, sigmoid, polynomial and linear kernel functions, and got 100% classification rate on all except for sigmoid kernel which had 52.08% classification rate. Decision tree technique developed using QUEST (Quick, Unbiased and Efficient Statistical Tree), C&RT (Classification and Regression Tree), and C5.0 algorithms were also applied to the same dataset and 100% classification rate was obtained, and 95.83% for Chi-square automatic interaction detection (CHAID) [23].…”

Section: Support Vector Machinementioning

confidence: 99%

“…Demetgul [23] used SVM to classify 9 fault states in a modular production system (MPS) using data obtained from eight sensors, and experimented with 4 different kernel functions namely RBF, sigmoid, polynomial and linear kernel functions, and got 100% classification rate on all except for sigmoid kernel which had 52.08% classification rate. Decision tree technique developed using QUEST (Quick, Unbiased and Efficient Statistical Tree), C&RT (Classification and Regression Tree), and C5.0 algorithms were also applied to the same dataset and 100% classification rate was obtained, and 95.83% for Chi-square automatic interaction detection (CHAID) [23]. Demetgul [23] concluded that SVM and decision tree algorithms are very effective monitoring and diagnostic tools for industrial production systems.…”

Section: Support Vector Machinementioning

confidence: 99%

“…Decision tree technique developed using QUEST (Quick, Unbiased and Efficient Statistical Tree), C&RT (Classification and Regression Tree), and C5.0 algorithms were also applied to the same dataset and 100% classification rate was obtained, and 95.83% for Chi-square automatic interaction detection (CHAID) [23]. Demetgul [23] concluded that SVM and decision tree algorithms are very effective monitoring and diagnostic tools for industrial production systems.…”

Section: Support Vector Machinementioning

confidence: 99%

See 3 more Smart Citations

A Review of Current Machine Learning Techniques Used in Manufacturing Diagnosis

Ademujimi

Brundage

Prabhu

2017

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Artificial intelligence applications are increasing due to advances in data collection systems, algorithms, and affordability of computing power. Within the manufacturing industry, machine learning algorithms are often used for improving manufacturing system fault diagnosis. This study focuses on a review of recent fault diagnosis applications in manufacturing that are based on several prominent machine learning algorithms. Papers published from 2007 to 2017 were reviewed and keywords were used to identify 20 articles spanning the most prominent machine learning algorithms. Most articles reviewed consisted of training data obtained from sensors attached to the equipment. The training of the machine learning algorithm consisted of designed experiments to simulate different faulty and normal processing conditions. The areas of application varied from wear of cutting tool in computer numeric control (CNC) machine, surface roughness fault, to wafer etching process in semiconductor manufacturing. In all cases, high fault classification rates were obtained. As the interest in smart manufacturing increases, this review serves to address one of the cornerstones of emerging production systems.

show abstract

Section: Support Vector Machinementioning

confidence: 99%

Section: Support Vector Machinementioning

confidence: 99%

Section: Support Vector Machinementioning

confidence: 99%

Section: Support Vector Machinementioning

confidence: 99%

See 2 more Smart Citations

A Review of Current Machine Learning Techniques Used in Manufacturing Diagnosis

Ademujimi

Brundage

Prabhu

2017

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

show abstract

“…Optimization of throughput [2], cycle time [2], output [33], and design [33,34], fault diagnosis of production systems [35] Warehouse [2,34], distribution center [33], food processing and packaging [36], pneumatic production system [35] Discrete event simulation [2], real data from a distribution center [33], simulation modular production system [mps] [35] Divide and conquer modeling approach and matrix-based method [2], heuristic decisionmaking rules [33], adjacency matrix, graph theory, optimization, decision support systems [34], Support Vector Machine [SVM] and Decision tree [35] Shorter distance and faster time compared to conventional control strategies [2], improvement in automation and throughput [33], 11 faults successfully diagnosed [35] the failure analysis of four components of a conveyor system based on a historical data collected from mining companies. The components included the drive unit, pulleys, idlers, and belts.…”

Section: Production Levelmentioning

confidence: 99%

A review of predictive monitoring approaches and algorithms for material handling systems

Siegel

Ardakani

Lee

et al. 2014

International Journal of Advanced Logistics

View full text Add to dashboard Cite

Material handling systems (MHS) include a wide range of systems and equipment that are used for handling, movement, storage and control of materials during manufacturing. MHS can represent a significant portion of the cost to make a product and there can be significant cost savings in manufacturing processes if unexpected failures of such systems can be avoided and maintenance costs lowered. This paper first surveys the common maintenance practices of MHS including three typical warehouse MHS e.g. automatic picking system (APS), goods to destination (GDS), and Erector. The survey from end users shows that the majority of the companies does not keep the record of their past failures or improve their maintenance practices after major failures occur. Even the end users with more advanced maintenance programs use more reactive maintenance approaches, as opposed to preventive maintenance approaches which have the potential to lower the chances of unexpected failures and costly repairs. The present paper also reviews the past works in predictive monitoring of MHS categorized as machine level, component level, and production level. Although there has not been a large body of research published around this topic, the existing works demonstrate the suitability of applying Prognostics and Health Management techniques for improving the reliability of such machines and reducing their downtime. The opportunity for MHS to move towards a condition-based maintenance approach is still a few years away, and many end-users of this equipment would have to first adopt a good preventive and reliably centered maintenance philosophy. Once that occurs, the OEM's would be the most likely candidates for further developing and implementing the research work and methods for MHS predictive monitoring.

show abstract

Support Vector Machines: A Review and Applications in Statistical Process Monitoring

Apsemidis¹,

Psarakis²

2020

Data Analysis and Applications 3

View full text Add to dashboard Cite

Fault diagnosis on production systems with support vector machine and decision trees algorithms

Cited by 46 publications

References 31 publications

A Review of Current Machine Learning Techniques Used in Manufacturing Diagnosis

A Review of Current Machine Learning Techniques Used in Manufacturing Diagnosis

A review of predictive monitoring approaches and algorithms for material handling systems

Support Vector Machines: A Review and Applications in Statistical Process Monitoring

Contact Info

Product

Resources

About