Standard

Learning about risk : Machine learning for risk assessment. / Paltrinieri, Nicola; Comfort, Louise; Reniers, Genserik.

In: Safety Science, Vol. 118, 01.10.2019, p. 475-486.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

Paltrinieri, N, Comfort, L & Reniers, G 2019, 'Learning about risk: Machine learning for risk assessment' Safety Science, vol. 118, pp. 475-486. https://doi.org/10.1016/j.ssci.2019.06.001

APA

Vancouver

Author

Paltrinieri, Nicola ; Comfort, Louise ; Reniers, Genserik. / Learning about risk : Machine learning for risk assessment. In: Safety Science. 2019 ; Vol. 118. pp. 475-486.

BibTeX

@article{1e7795d30462499bb152021ea27bc43f,
title = "Learning about risk: Machine learning for risk assessment",
abstract = "Risk assessment has a primary role in safety-critical industries. However, it faces a series of overall challenges, partially related to technology advancements and increasing needs. There is currently a call for continuous risk assessment, improvement in learning past lessons and definition of techniques to process relevant data, which are to be coupled with adequate capability to deal with unexpected events and provide the right support to enable risk management. Through this work, we suggest a risk assessment approach based on machine learning. In particular, a deep neural network (DNN) model is developed and tested for a drive-off scenario involving an Oil & Gas drilling rig. Results show reasonable accuracy for DNN predictions and general suitability to (partially) overcome risk assessment challenges. Nevertheless, intrinsic model limitations should be taken into account and appropriate model selection and customization should be carefully carried out to deliver appropriate support for safety-related decision-making.",
keywords = "Deep learning, Dynamic risk analysis, Machine learning, Risk assessment",
author = "Nicola Paltrinieri and Louise Comfort and Genserik Reniers",
year = "2019",
month = "10",
day = "1",
doi = "10.1016/j.ssci.2019.06.001",
language = "English",
volume = "118",
pages = "475--486",
journal = "Safety Science",
issn = "0925-7535",
publisher = "Elsevier",

}

RIS

TY - JOUR

T1 - Learning about risk

T2 - Safety Science

AU - Paltrinieri, Nicola

AU - Comfort, Louise

AU - Reniers, Genserik

PY - 2019/10/1

Y1 - 2019/10/1

N2 - Risk assessment has a primary role in safety-critical industries. However, it faces a series of overall challenges, partially related to technology advancements and increasing needs. There is currently a call for continuous risk assessment, improvement in learning past lessons and definition of techniques to process relevant data, which are to be coupled with adequate capability to deal with unexpected events and provide the right support to enable risk management. Through this work, we suggest a risk assessment approach based on machine learning. In particular, a deep neural network (DNN) model is developed and tested for a drive-off scenario involving an Oil & Gas drilling rig. Results show reasonable accuracy for DNN predictions and general suitability to (partially) overcome risk assessment challenges. Nevertheless, intrinsic model limitations should be taken into account and appropriate model selection and customization should be carefully carried out to deliver appropriate support for safety-related decision-making.

AB - Risk assessment has a primary role in safety-critical industries. However, it faces a series of overall challenges, partially related to technology advancements and increasing needs. There is currently a call for continuous risk assessment, improvement in learning past lessons and definition of techniques to process relevant data, which are to be coupled with adequate capability to deal with unexpected events and provide the right support to enable risk management. Through this work, we suggest a risk assessment approach based on machine learning. In particular, a deep neural network (DNN) model is developed and tested for a drive-off scenario involving an Oil & Gas drilling rig. Results show reasonable accuracy for DNN predictions and general suitability to (partially) overcome risk assessment challenges. Nevertheless, intrinsic model limitations should be taken into account and appropriate model selection and customization should be carefully carried out to deliver appropriate support for safety-related decision-making.

KW - Deep learning

KW - Dynamic risk analysis

KW - Machine learning

KW - Risk assessment

UR - http://www.scopus.com/inward/record.url?scp=85066620513&partnerID=8YFLogxK

U2 - 10.1016/j.ssci.2019.06.001

DO - 10.1016/j.ssci.2019.06.001

M3 - Article

VL - 118

SP - 475

EP - 486

JO - Safety Science

JF - Safety Science

SN - 0925-7535

ER -

ID: 55930456