An integrated method for improving risk analysis using human factors methods and virtual reality [Elektronische Ressource] / von: Waleed Salem

An Integrated Method for Improving Risk Analysis Using Human Factors
Methods and Virtual Reality


Dissertation

zur Erlangung des akademischen Grades

Doktoringenieurin oder Doktoringenieur (Dr.-Ing.)



angenommen durch die Fakultät für Informatik
der Otto-von-Guericke-Universität Magdeburg


von: M. Sc. Waleed Salem
geb. am: 02.09.1974 in: Ramtha




Gutachterinnen/Gutachter:

Prof. Dr. Andreas Nürnberger
Prof. Dr. Tom Kontogiannis
Prof. Dr. Teodor Winkler



Ort und Datum des Promotionskolloquiums: Magdeburg, 19.05.2009
I


II

Abstract......
Increasing accident rates and system losses in different industrial sectors endanger
safety, threaten economic growth and cause pollution damages. A major cause for these
accidents and losses in the chemical process industry is the human error which
contributes with a range of 60-90% in the development of these accidents. Conventional
safety and risk analysis methods focus mainly on describing technological malfunctions
and lack a systematic consideration of the human impact, i.e., the human error, on the
process under consideration. These methods lack also a systematic utilisation of existing
supporting and enabling technologies, e.g. virtual reality.
In this thesis, an integrated method for improving risk analysis is introduced. This
method is developed by utilising human factors knowledge for a better inclusion of
human impact in the risk analysis. It also utilises virtual reality as enabling technology
that can be used as a medium for running relevant safety scenarios. The utilisation of
human factors methods and virtual reality is based on an end user oriented approach of
data collection, review and validation from the selected industrial domain (Chemical
Process Industry).
III

IV

Zusammenfassung
Steigende Unfallzahlen und Ausfälle von technischen Systemen in verschiedenen
Industriebranchen gefährden die Sicherheit, bedrohen das Wirtschaftswachstum und
tragen zur Umweltverschmutzung bei. Eine wesentliche Ursache für diese Unfälle und
Verluste in der chemischen Prozessindustrie ist menschliches Versagen, das mit einem
Anteil von 60-90% zu Entwicklung dieser Unfälle beiträgt. Konventionelle Methoden
der Sicherheits- und Risikoanalyse fokussieren hauptsächlich auf die Beschreibung
technischer Störungen und Abweichungen und ihnen fehlt eine systematische
Berücksichtigung der menschlichen Einflüsse (d. h., der menschliche Fehler) im
betrachteten Prozess. Des Weiteren fehlt diesen Methoden eine gezielte Nutzung
vorhandener unterstützender Technologien, z. B., Virtual Reality.
In dieser Dissertation wird eine integrierte Methode zur Verbesserung der Risikoanalyse
eingeführt. Diese Methode basiert auf eine Nutzung der Erkenntnisse der menschlichen
Faktoren (Human Factors) zur verbesserten Einbeziehung der menschlichen Einflüsse in
der Risikoanalyse. Die Methode nutzt auch die virtuelle Realität als technologische
Umgebung zur Durchführung und Erprobung relevanter Sicherheitsszenarien. Die
Nutzung der menschlichen Faktoren und die virtuelle Realität basieren auf einem
Endnutzer-orientierten Verfahren zur Datenerhebung, Revision und Validierung aus der
ausgewählten Industriebranche (chemische Prozessindustrie).
V

VI


Table of contents
Abstract...... ...................................................................................................................... III
Zusammenfassung ............ V
Table of contents ........... VII
List of abbreviations ....................................................................................................... XI
List of figures ................ XII
List of tables .................................................................................................................XIV
1 Introduction ................... 1
1.1 Motivation and current situation ......................................................................... 2
1.2 Goal definition ..................................... 6
1.3 Description of the integrated method .. 9
1.4 Contributions ..................................................................... 11
1.5 Research domain ............................................................... 11
1.6 Structure of the document ................................................. 12
2 State-of-the-art and related work ................................................ 15
2.1 Risks, risk management and risk analysis ......................... 16
2.1.1 Definitions ............................................................................................. 16
2.1.2 Types of risks ........................ 17
2.1.3 Risk management process ..................................... 19
2.1.4 Risk analysis ......................................................................................... 20
2.1.5 Rlysis methods (RA methods) ................... 21
2.1.6 Quantitative risk analysis (QRA) .......................... 22
2.1.6.1 Identify hazards and potential hazardous events .................... 22
2.1.6.2 Development of top events into incident scenarios and
estimation of frequencies ........................................................ 24
2.1.6.3 Assessment of consequences and calculation of potential
loss from incident scenarios .................... 26
2.1.7 Summary of Section 2.1 ........................................................................ 26
2.2 Human Reliability (HR) and Human Factors (HF) ........... 27
2.2.1 Definitions ............................................................. 27
2.2.2 Human Reliability Analysis (HRA) ...................................................... 28
2.2.3 1st Generation of HRA methods and techniques .................................. 30
st2.2.3.1 Limitations and shortcomings of 1 generation methods ....... 30
nd2.2.4 2 Generation of HRA methods and techniques .. 32
nd2.2.4.1 Limitations and shortcomings of 2 generation methods ...... 33
2.2.5 Human factors methods and techniques ................................................ 34
2.2.5.1 Task Analysis (TA) ................................................................. 36
2.2.5.2 Root-Cause Analysis ............................... 38
2.2.5.3 Performance shaping factors (PSFs) ....... 38
VII
2.2.6 Summary of Section 2.2 ........................................................................ 39
2.3 Virtual Reality (VR) .......................................................... 39
2.3.1 Definitions ............................. 40
2.3.2 Introduction to Virtual Reality .............................................................. 41
2.3.2.1 Immersive and Non-Immersive Virtual Reality ..................... 41
2.3.2.2 Layout of a VR system ........................... 42
2.3.2.3 Reasons for selecting VR as enabler to support RA ............... 44
2.3.2.4 Areas of application of Virtual Reality ................................... 46
2.3.3 Results of surveys on VR ...................................... 47
2.3.3.1 External surveys ...... 48
2.3.3.2 Internal surveys ....................................... 52
2.3.4 Summary of Section 2.3 ........................................ 58
2.4 Summary and conclusions ................. 58
3 A HF methodology for supporting risk analysis ......................................................... 61
3.1 Definitions ......................................................................................................... 62
3.2 Preamble ............ 64
3.3 Error causation model ....................................................................................... 66
3.4 The methodology ............................... 67
3.4.1 A model of cognitive functions ............................. 68
3.4.2 Error modes ........................................................................................... 70
3.4.2.1 External error modes ............................... 70
3.4.2.2 Cognitive (internal) error modes ............................................. 72
3.4.3 Performance mechanisms ..................................... 73
3.4.3.1 Identification and Interpretation ............. 75
3.4.3.2 Decision-making (choice of goals) and Planning ................... 76
3.4.4 Performance conditions ......................................................................... 76
3.4.5 Cognitive Error Causes ......... 79
3.4.5.1 Error causes in identification and interpretation ..................... 79
3.4.5.2 Error causes in decision-making ............................................. 81
3.4.5.3 Error causes in planning.......................... 81
3.4.6 Error detection and recovery ................................................................. 83
3.5 Applying the methodology to risk analysis ....................... 87
3.6 Summary ........................................... 91
4 Functional VR requirements and specifications ......................................................... 95
4.1 Definitions .................................................................