Comparing capability of scenario hazard identification methods by the PIC (Plant-People-Procedure Interaction Contribution) network metric

Benjamin J. Seligmann, Jiayi Zhao, Sophia G. Marmara, Tayla C. Corbett, Michael Small, Maureen Hassall, Jesse T. Boadle

Research output: Contribution to journalArticle

Abstract

Comparing the results of hazard identification (HAZID) methods is a complex task, but the question that drives this activity is vitally important: which HAZID method should be used to best identify an accident scenario? Despite many efforts to address this, effective metrics do not yet readily exist for clearly comparing HAZID results for a particular scenario. The complexity of socio-technical systems is often cited as a key factor that limits effective scenario identification, calling into question traditional HAZID efforts. Motivated by the observation that interactions between multiple component types, such as People, Plant and Procedures (P3), often significantly contribute to major process system accidents, being an expression of the complexity of the system, a novel, precise, network topology-based metric for calculating the contribution of P3 Interactions to accident scenarios is presented. This metric, called the P3 Interaction Contribution (PIC), is intended to be used for comparing the HAZID results. An illustrative example of using the PIC for HAZID comparison is included, whereby Failure Mode and Effects Analysis (FMEA), Blended Hazard Identification methodology (BLHAZID) and Systems Theoretic Process Analysis (STPA) were each applied to a heat exchanger start-up operation. The results show initial promise that the PIC can be effective as a HAZID comparison tool. The main outputs of this paper are the presentation of the PIC calculation process and the method for applying the PIC to HAZID results. The paper concludes with recommendations for further experimental work to explore the validation and assess the true value of the PIC.

Original languageEnglish
Pages (from-to)116-129
Number of pages14
JournalSafety Science
Volume112
DOIs
Publication statusPublished - 1 Feb 2019

Fingerprint

Hazards
scenario
interaction
Accidents
accident
process analysis
systems analysis
Failure modes
Heat exchangers
heat
Hot Temperature
Topology
methodology
Values

Cite this

Seligmann, Benjamin J. ; Zhao, Jiayi ; Marmara, Sophia G. ; Corbett, Tayla C. ; Small, Michael ; Hassall, Maureen ; Boadle, Jesse T. / Comparing capability of scenario hazard identification methods by the PIC (Plant-People-Procedure Interaction Contribution) network metric. In: Safety Science. 2019 ; Vol. 112. pp. 116-129.
@article{3615127e51784f7da13f9d503a997651,
title = "Comparing capability of scenario hazard identification methods by the PIC (Plant-People-Procedure Interaction Contribution) network metric",
abstract = "Comparing the results of hazard identification (HAZID) methods is a complex task, but the question that drives this activity is vitally important: which HAZID method should be used to best identify an accident scenario? Despite many efforts to address this, effective metrics do not yet readily exist for clearly comparing HAZID results for a particular scenario. The complexity of socio-technical systems is often cited as a key factor that limits effective scenario identification, calling into question traditional HAZID efforts. Motivated by the observation that interactions between multiple component types, such as People, Plant and Procedures (P3), often significantly contribute to major process system accidents, being an expression of the complexity of the system, a novel, precise, network topology-based metric for calculating the contribution of P3 Interactions to accident scenarios is presented. This metric, called the P3 Interaction Contribution (PIC), is intended to be used for comparing the HAZID results. An illustrative example of using the PIC for HAZID comparison is included, whereby Failure Mode and Effects Analysis (FMEA), Blended Hazard Identification methodology (BLHAZID) and Systems Theoretic Process Analysis (STPA) were each applied to a heat exchanger start-up operation. The results show initial promise that the PIC can be effective as a HAZID comparison tool. The main outputs of this paper are the presentation of the PIC calculation process and the method for applying the PIC to HAZID results. The paper concludes with recommendations for further experimental work to explore the validation and assess the true value of the PIC.",
keywords = "Accident, Complexity, Hazard identification, Network, Socio-technical",
author = "Seligmann, {Benjamin J.} and Jiayi Zhao and Marmara, {Sophia G.} and Corbett, {Tayla C.} and Michael Small and Maureen Hassall and Boadle, {Jesse T.}",
year = "2019",
month = "2",
day = "1",
doi = "10.1016/j.ssci.2018.10.019",
language = "English",
volume = "112",
pages = "116--129",
journal = "Safety Science",
issn = "0925-7535",
publisher = "Elsevier",

}

Comparing capability of scenario hazard identification methods by the PIC (Plant-People-Procedure Interaction Contribution) network metric. / Seligmann, Benjamin J.; Zhao, Jiayi; Marmara, Sophia G.; Corbett, Tayla C.; Small, Michael; Hassall, Maureen; Boadle, Jesse T.

In: Safety Science, Vol. 112, 01.02.2019, p. 116-129.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Comparing capability of scenario hazard identification methods by the PIC (Plant-People-Procedure Interaction Contribution) network metric

AU - Seligmann, Benjamin J.

AU - Zhao, Jiayi

AU - Marmara, Sophia G.

AU - Corbett, Tayla C.

AU - Small, Michael

AU - Hassall, Maureen

AU - Boadle, Jesse T.

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Comparing the results of hazard identification (HAZID) methods is a complex task, but the question that drives this activity is vitally important: which HAZID method should be used to best identify an accident scenario? Despite many efforts to address this, effective metrics do not yet readily exist for clearly comparing HAZID results for a particular scenario. The complexity of socio-technical systems is often cited as a key factor that limits effective scenario identification, calling into question traditional HAZID efforts. Motivated by the observation that interactions between multiple component types, such as People, Plant and Procedures (P3), often significantly contribute to major process system accidents, being an expression of the complexity of the system, a novel, precise, network topology-based metric for calculating the contribution of P3 Interactions to accident scenarios is presented. This metric, called the P3 Interaction Contribution (PIC), is intended to be used for comparing the HAZID results. An illustrative example of using the PIC for HAZID comparison is included, whereby Failure Mode and Effects Analysis (FMEA), Blended Hazard Identification methodology (BLHAZID) and Systems Theoretic Process Analysis (STPA) were each applied to a heat exchanger start-up operation. The results show initial promise that the PIC can be effective as a HAZID comparison tool. The main outputs of this paper are the presentation of the PIC calculation process and the method for applying the PIC to HAZID results. The paper concludes with recommendations for further experimental work to explore the validation and assess the true value of the PIC.

AB - Comparing the results of hazard identification (HAZID) methods is a complex task, but the question that drives this activity is vitally important: which HAZID method should be used to best identify an accident scenario? Despite many efforts to address this, effective metrics do not yet readily exist for clearly comparing HAZID results for a particular scenario. The complexity of socio-technical systems is often cited as a key factor that limits effective scenario identification, calling into question traditional HAZID efforts. Motivated by the observation that interactions between multiple component types, such as People, Plant and Procedures (P3), often significantly contribute to major process system accidents, being an expression of the complexity of the system, a novel, precise, network topology-based metric for calculating the contribution of P3 Interactions to accident scenarios is presented. This metric, called the P3 Interaction Contribution (PIC), is intended to be used for comparing the HAZID results. An illustrative example of using the PIC for HAZID comparison is included, whereby Failure Mode and Effects Analysis (FMEA), Blended Hazard Identification methodology (BLHAZID) and Systems Theoretic Process Analysis (STPA) were each applied to a heat exchanger start-up operation. The results show initial promise that the PIC can be effective as a HAZID comparison tool. The main outputs of this paper are the presentation of the PIC calculation process and the method for applying the PIC to HAZID results. The paper concludes with recommendations for further experimental work to explore the validation and assess the true value of the PIC.

KW - Accident

KW - Complexity

KW - Hazard identification

KW - Network

KW - Socio-technical

UR - http://www.scopus.com/inward/record.url?scp=85055751150&partnerID=8YFLogxK

U2 - 10.1016/j.ssci.2018.10.019

DO - 10.1016/j.ssci.2018.10.019

M3 - Article

VL - 112

SP - 116

EP - 129

JO - Safety Science

JF - Safety Science

SN - 0925-7535

ER -