TY - JOUR
T1 - A polynomial neural network approach for improving risk assessment and industrial safety
AU - Guzman-Urbina, Alexander
AU - Aoyama, Atsushi
AU - Choi, Eugene
N1 - Funding Information:
Acknowledgment. The authors would like to extend his gratitude to the Otsuka Toshimi Scholarship Foundation for contribution to the realization of this research paper.
Publisher Copyright:
© 2018, ICIC International. All rights reserved.
PY - 2018/2
Y1 - 2018/2
N2 - The estimation of the magnitude of accidental risks is one of the key elements for decision making in safety of industrial infrastructure. However, modeling industrial accidents involves very complex structures of calculation due to the considerable number of variables and large degree of uncertainty. Therefore, this paper aims to introduce a polynomial neural network approach to estimating accidental risks by using the grouping method of data handling (GMDH), which is able to deal efficiently with the complexity of the data by optimizing the configuration of the risk assessment model. The findings of this study show that accidental risk values could be improved using a GMDH-type neural network in contrast with the traditional methods by increasing the precision of risk estimation. Additional to this contribution, this study highlights the sensible and critical parameters of GMDH-type neural network for risk assessment.
AB - The estimation of the magnitude of accidental risks is one of the key elements for decision making in safety of industrial infrastructure. However, modeling industrial accidents involves very complex structures of calculation due to the considerable number of variables and large degree of uncertainty. Therefore, this paper aims to introduce a polynomial neural network approach to estimating accidental risks by using the grouping method of data handling (GMDH), which is able to deal efficiently with the complexity of the data by optimizing the configuration of the risk assessment model. The findings of this study show that accidental risk values could be improved using a GMDH-type neural network in contrast with the traditional methods by increasing the precision of risk estimation. Additional to this contribution, this study highlights the sensible and critical parameters of GMDH-type neural network for risk assessment.
KW - Artificial intelligence
KW - GMDH
KW - Industrial safety
KW - Polynomial neural network
KW - Risk assessment
UR - http://www.scopus.com/inward/record.url?scp=85040786268&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85040786268&partnerID=8YFLogxK
U2 - 10.24507/icicel.12.02.97
DO - 10.24507/icicel.12.02.97
M3 - Article
AN - SCOPUS:85040786268
VL - 12
SP - 97
EP - 107
JO - ICIC Express Letters
JF - ICIC Express Letters
SN - 1881-803X
IS - 2
ER -