Faculty of Science and Technology
MASTER’S THESIS
Study program/ Specialization:
Offshore Technology/
Industrial Asset Management
Spring semester, 2015
Open / Restricted access
Writer:
Andika Rachman ………
(Writer’s signature)
Faculty supervisor: R.M. Chandima Ratnayake
External supervisor(s): Arnaud Barre
Thesis title:
On the Use of Historical Failures Patterns to Confront the Unforeseen
Credits (ECTS): 30
Key words:
unforeseen failure black swan
accident
Pages: 84 + enclosure: 16
Stavanger, 15
thJune 2015
Unforeseen Events
Unknown unknowns
Unknown knowns
Events that are believed to have negligilble probability
Probability-based
thinking Knowledge
dimension Black swans
+ +
Preoccupation with failure
Reluctance to simplify
Sensitivity operationsto Commitment
to resilience Deference to
expertise
Unexpected Event (Unsafe Act)
Normally Functioning
System
Unwanted Consequences
Direction of Causality
Direction of Reasoning
Effect Cause
If we can see what the effect is...
Then we can find what the
cause is
Pipeline washing Water entered MIC tank through leaky
valve
Contamination of MIC with water caused runaway
reaction
High temperature caused the bursting
of the tank casing
Boundary to unacceptable
workload
Boundary to economic failure
Resulting perceived boundary of acceptable performance, i.e.
safety boundary defined by work practices
Boundary of functionally acceptable performance, i.e. the
real safety boundary
Sy stem
Management effort toward efficiency
Gradient toward
less effort
Accident Counter forces represented by safety
management system
Space of possibilities; day-to-day
operation; degrees of freedom to be resolves according subjective
preferences
Movement towards degrading safety defenses
Controller
Controlled Process
Actuator Sensor
Safety Management Process (Internal Variability)
Leading Indicators Environment (External Variability)
Safety Performance Desired Safety
Level Control
Signal
Feedback
Pressure Transmitter (PT)
• Send signal to HCM to indicate the flowlines pressure
HIPPS Control Module (HCM)
• Process the information from PT and send decisions signal to SDV
Shutdown Valve (SDV)
• Close or open the
flow based on
signal from HCM
HIPPS
Operating Company Petroleum Safety AuthorityOffshore Installation Manager
Operation Team Leader Regulations
Standards Certif ications
Accident Reports Saf ety Rev iews Whistleblowers
Saf ety Policy Standards Resources
Operation Reports Audit and Change Reports
Work Instructions Operating Manuals
Saf ety Policy Standards Resources
Operation Reports Problem Reports Change Requests
Problem Reports Operation Reports Change Requests
Operators
Alarms Pressure Trends Valv e Status Testing
Monitoring Correctiv e Actions Prev entiv e Actions
MCS
HCM Operator
SDV (Close/Open)
Alarm SDV Status
Signal to conduct PST PST
command
SDV Status Test result
interpretation
Deactivate solenoid valve
SDV Status Reactivate
solenoid valve
Process ModelOperator
· Standard procedure
· Training
· Safety policy
· Testing acceptance criteria
Alarm SDV Status
Signal to conduct PST Test command
SDV
SDV Status Test result
interpretation
Deactivate
solenoid valve SDV Status
Process ModelMCSTest command
· Send signal to conduct PST SDV Status Indicator
· Open
· Close
HCM Test command
· Deactivate solenoid valve
· Reactivate solenoid valve before the SDV reaches fully closed condition
Reactivate
solenoid valve
MCS
HCM Operator
The testing result is judged as acceptable when it is actually not
MCS provides false indication of SDV status
SCM provides false indication of SDV status
SDV Operator misinteprets output signal from MCS
No alarm is provided when SDV fails to close
MCS
HCM Operator
SDV is fully closed during partial stroke testing
MCS executes different command from the one that operator has given Operator sends wrong command to MCS
SDV
Failure of hydraulic lineFailure of the solenoid valves
Failure of electrical line Failure of electrical line