A multi-layer neural network model for detecting changes in the process mean

doi:10.1016/0360-8352(94)00024-H

Computers & Industrial Engineering

Volume 28, Issue 1, January 1995, Pages 51-61

https://doi.org/10.1016/0360-8352(94)00024-H Get rights and content

Abstract

Control charts are an important tool in statistical process control (SPC). They are useful in determining whether a process is behaving as intended or if there are some unnatural causes of variation. In this paper we present an alternative approach to SPC using artificial neural network technology and compare its performance with that of the combined Shewhart-CUSUM schemes. The neural network developed here is concerned with the detection of gradual trends and sudden shifts in the process mean. The performance of the neural network was evaluated by estimating the average run lengths (ARL's) using simulation. An extensive comparison shows that the proposed approach has 20–40% faster detection of small process changes than the combined Shewhart-CUSUM control schemes. The neural network approach presented in this paper appears to offer a competitive alternative to existing control schemes.

References (16)

G.A. Pugh
Synthetic neural networks for process control
Computers ind. Engng
(1989)
G.A. Pugh
A comparison of neural networks to SPC charts
Computers ind. Engng
(1991)
H.B. Hwarng et al.
Back-propagation pattern recognizers for x̄
Computers ind. Engng
(1993)
Western Electric Company
E.L. Grant et al.
Statistical Quality Control
(1988)
C.W. Champ et al.
Exact results for Shewhart control charts with supplementary runs rules
Technometrics
(1987)
E.S. Page
Continuous inspection schemes
Biometrika
(1954)
J.M. Lucas
A modified V mask control scheme
Technometrics
(1973)

There are more references available in the full text version of this article.

Cited by (136)

Nanoscale electronic synapses for neuromorphic computing
2022, Intelligent Nanotechnology: Merging Nanoscience and Artificial Intelligence
With the increasing density of transistors, the bottleneck of data transmission and distribution will become insurmountable, ultimately limiting the computing power. There is a great chance to solve this problem when an electronic synaptic system inspired by the biological brain is combined with neuromorphic computing, which has strong computing power and high energy efficiency. Herein, we offer a review of the emerging electronic synaptic techniques, including the operating mechanisms of electron synapses and the realization of neuromorphic computing with electron synapses. This review may provide clues and enlightenment for the working mechanism, material selection, and device design of electronic synapses for neuromorphic computing.
A combined machine learning algorithms and DEA method for measuring and predicting the efficiency of Chinese manufacturing listed companies
2021, Journal of Management Science and Engineering
Data Envelopment Analysis (DEA) is a linear programming methodology for measuring the efficiency of Decision Making Units (DMUs) to improve organizational performance in the private and public sectors. However, if a new DMU needs to be known its efficiency score, the DEA analysis would have to be re-conducted, especially nowadays, datasets from many fields have been growing rapidly in the real world, which will need a huge amount of computation. Following the previous studies, this paper aims to establish a linkage between the DEA method and machine learning (ML) algorithms, and proposes an alternative way that combines DEA with ML (ML-DEA) algorithms to measure and predict the DEA efficiency of DMUs. Four ML-DEA algorithms are discussed, namely DEA-CCR model combined with back-propagation neural network (BPNN-DEA), with genetic algorithm (GA) integrated with back-propagation neural network (GANN-DEA), with support vector machines (SVM-DEA), and with improved support vector machines (ISVM-DEA), respectively. To illustrate the applicability of above models, the performance of Chinese manufacturing listed companies in 2016 is measured, predicted and compared with the DEA efficiency scores obtained by the DEA-CCR model. The empirical results show that the average accuracy of the predicted efficiency of DMUs is about 94%, and the comprehensive performance order of four ML-DEA algorithms ranked from good to poor is GANN-DEA, BPNN-DEA, ISVM-DEA, and SVM-DEA.
Modeling the total heat transfer coefficient of a nuclear research reactor cooling system by different methods
2021, Case Studies in Thermal Engineering
Citation Excerpt :
These networks consist of 3 layers. These layers are the input layer, the hidden layer, and the output layer [49]. The input and output layers are chosen among the parameters that affect the solution.
In this work, the thermal model of Triga Mark-II Nuclear research reactor working at Istanbul Technical University (ITU) Energy Institute was created. Thermal model includes mathematical equations expressing the change of water inlet-outlet temperatures of reactor tank, heat exchanger and cooling tower depending on time. By using the thermal model, the mass flow rates of the first and second cooling circuits that can affect the power of the reactor's cooling system were examined. At the same time, an equation based on NTU was generated to estimate the total heat transfer coefficient. To add originality and innovation to the study, the total heat transfer coefficient was modeled using Machine Learning Algorithms (Multilayer Perceptron, Support Vector Machine, M5P Model Tree). Model data obtained by utilizing the thermal model and machine learning were compared. As a result, the mass flow rate at which the reactor works efficiently was determined as 6.57–8.5 kg/h for the 1st cooling circuit and 11.5–13 kg/h for the 2nd cooling circuit. SVM (Puk Kernel) is the method with the least error (RMSE: 0.223) among NTU equation, Thermal model and Machine learning algorithms.
Quality Control Using Convolutional Neural Networks Applied to Samples of Very Small Size
2023, Stochastics and Quality Control
Application of Multilayer Neural Network in Sports Psychology
2022, Scientific Programming
Extraction and organization of statistical distribution functions for simulation of variations and patterns in the variability control charts
2021, Journal of Statistical Computation and Simulation

View all citing articles on Scopus

View full text

A multi-layer neural network model for detecting changes in the process mean

Abstract

Computers ind. Engng

Computers ind. Engng

Computers ind. Engng

Statistical Quality Control

Exact results for Shewhart control charts with supplementary runs rules

Technometrics

Continuous inspection schemes

Biometrika

A modified V mask control scheme

Technometrics