From Process Safety Management to Operational
Excellence
February 28, 2008
Ian Sutton
Introduction
Process Industries
– AMEC Paragon
– Design and Operations
Various books / ebooks
– Process Risk Management
– Procedures and Training in the Process
Industries
– Management of Change
– Fault Tree Analysis
– www.stb07.com
Writing Operational Excellence and
Incident Investigation and Analysis
From PSM to OE. Ian Sutton. 2/28/08. 2
Goals of the Presentation
1. What's the problem?
2. What's the solution?
3. What's the cost?
From PSM to OE. Ian Sutton. 2/28/08. 3
What's the Problem / Issue?
PSM is mature (probably similar for safety
cases). 20 years old. This is a compliment.
Many facilities now have a good functioning
program
– There is always room for improvement
– Some companies / industries still have major issues
– Nevertheless, many companies have "arrived"
– They "survive an audit"
– Not much development - they are maintaining what
they have
PSM is not structured to make money
– Regulatory compliance
– Safety
What are the development options for PSM?
From PSM to OE. Ian Sutton. 2/28/08. 4
What's the Solution?
PSM is not the only management program implemented in
the last twenty years
Others include ISO 9000, safety culture, incident analysis,
risk-based inspection, system availability, security, six
sigma and behavior-based safety
Integrate the best features of these programs – create a
synergy
Search for common themes
Thereby achieve operational excellence
From PSM to OE. Ian Sutton. 2/28/08. 5
What's the Cost?
Need a practical program
Speedy implementation
Build on the funds already spent
From PSM to OE. Ian Sutton. 2/28/08. 6
Common Themes
1. Internalize Risk Perception
2. Understand Engineering Standards
3. Learn from Audits and Incident Investigations
4. Restructure Hazards Analyses
5. Learn from Events Elsewhere
The Elevator Talk (30 – 45 seconds):
All employees, managers, designers and contract
workers must understand the systems for which
they are responsible. Hence, education - as distinct
from training - is both a need and an opportunity.
From PSM to OE. Ian Sutton. 2/28/08. 7
Seven Sections: Merging other Programs
1. Defining Operational Excellence
2. Historical Development of HSE
3. Process Safety Management
4. Safety Culture
5. Incident Analysis
6. System Availability and
Troubleshooting
7. Creating Operational Excellence
From PSM to OE. Ian Sutton. 2/28/08. 8
Section 1 – Defining Operational Excellence
From PSM to OE. Ian Sutton. 2/28/08. 9
Operational Integrity Management
Operational
PSM Integrity ISO 9000 /14000
Management
Si
x
E Si
S gm
H a
RAM
From PSM to OE. Ian Sutton. 2/28/08. 10
Operational Excellence
From PSM to OE. Ian Sutton. 2/28/08. 11
Simplify
From PSM to OE. Ian Sutton. 2/28/08. 12
Historical Development
From PSM to OE. Ian Sutton. 2/28/08. 13
Returns Become More Difficult
From PSM to OE. Ian Sutton. 2/28/08. 14
Section 3 – Process Safety Management
From PSM to OE. Ian Sutton. 2/28/08. 15
Types of Safety
From PSM to OE. Ian Sutton. 2/28/08. 16
Occupational and Process Safety
BP’s executive management tracked the trends in BP’s
personal safety metrics, and they understood that BP’s
performance in this regard was both better than industry
averages and consistently improving. Based upon these
trends, BP’s executive management believed that the
focus on metrics such as OSHA recordables and the
implementation of the Group-wide driving standard were
largely successful. With respect to personal safety, that
focus evidently was effective. BP’s executive
management, however, mistakenly believed that injury
rates, such as days away from work case frequency
and recordable injury frequency, were indicators of
acceptable process safety performance.
From PSM to OE. Ian Sutton. 2/28/08. 17
Care with the Incident Pyramid
From PSM to OE. Ian Sutton. 2/28/08. 18
Regulatory Strategies
From PSM to OE. Ian Sutton. 2/28/08. 19
20 Years Ago
Process Safety Management (PSM) about 20
years old
– OSHA 29 CFR 1910.119
– EPA RMP
– API RP 75
– AIChE
– Many internal standards
High degree of similarity
Not really new
Regulations forced completion quickly
Gave support to mid-level managers and
engineers
From PSM to OE. Ian Sutton. 2/28/08. 20
Elements of OE
Include
– Safety culture
– Incident analysis
– System availability and troubleshooting
Exclude
– ISO 9000/14000
– Health, safety and environmental programs
– Six sigma
– Security
– Outsourcing
– Enterprise solutions
– Just in time management
From PSM to OE. Ian Sutton. 2/28/08. 21
Section 2 – Historical Development
Most discoveries are made regularly every fifteen years.
G.B. Shaw (1856-1950)
Those who cannot learn from history are doomed to
repeat it.
George Santayana (1863 – 1952)
From PSM to OE. Ian Sutton. 2/28/08. 22
Why Study the History of HSE?
By reviewing historical events we
see how they might happen again.
Helps develop an understanding of
overall trends, thus making efficient
use of investment.
Creates and retains institutional and
organizational memory.
Fits with the theme of using existing
programs and initiatives.
From PSM to OE. Ian Sutton. 2/28/08. 23
Elements of PSM (Regulatory)
1. Employee Participation 8. Mechanical Integrity
2. Process Safety Information 9. Hot Work
3. Process Hazards Analysis 10. Management of Change
4. Operating Procedures 11. Incident Investigation
5. Training 12. Emergency Planning and
6. Contractors Response
7. Prestartup Safety Review 13. Compliance Audits
14. Trade Secrets
From PSM to OE. Ian Sutton. 2/28/08. 24
Sixteen Elements
1. Participation, Leadership and 9. Occupational and Behavior-
Accountability Based Safety
2. Information and Quality 10. Operations and Maintenance
Management 11. Emergency Planning and
Response
3. Hazards Analysis
12. Incident Investigation and
4. Management of Change Analysis
5. Prestartup Safety Review 13. Human Factors
6. Operating Procedures 14. System Availability and
7. Training and Education Troubleshooting
8. Equipment and Instrument 15. Communication with the
Public and Security
Integrity
16. Audits and Assessments
From PSM to OE. Ian Sutton. 2/28/08. 25
Features of PSM
Participation, leadership and Utilization of the principles of
accountability risk management
– Linked to behavioral safety – Understand the risk equation
concepts – Define "acceptable risk"
– Responsible for one another Management and control
– "If you see it, you own it"
– Not fundamentally about safety
– Creative
– Define control
Non-prescriptive / – Attain control
performance-based – Respond to loss of control
– Goal-driven – Elements are integrated
– Cannot be "in compliance"
– Details vary a lot
– Measure results in terms of
value to client or customer
From PSM to OE. Ian Sutton. 2/28/08. 26
Section 4 – Safety Culture
Organizations don’t have memories — only people do.
Trevor Kletz
From PSM to OE. Ian Sutton. 2/28/08. 27
Lessons from Longford / Warning Flags
. . . when we extend the causal network < for
an accident > far enough, market forces and
cost cutting pressures are almost invariably
implicated.
Andrew Hopkins
From PSM to OE. Ian Sutton. 2/28/08. 28
The Titanic (Brander)
Most of the discussion revolves around specific problems.
There was the lack of sufficient lifeboats (enough for at most
1200 on a ship carrying 2200). There was the steaming ahead
at full-speed despite various warnings about the ice-field.
There was the lack of binoculars for the lookout. There were
the poor procedures with the new invention, the wireless (not
all warnings sent to the ship reached the bridge, and a nearby
ship, the operator abed, missed Titanic's SOS). Very recently,
from recovered wreckage, "Popular Science" claimed the hull
was particularly brittle even for the metallurgy of the time. (A
claim now debunked.) Each has at one time or another been
put forward as "THE reason the Titanic sank".
What gets far less comment is that most of the problems all
came from a larger, systemic problem: the owners and
operators of steamships had for five decades taken larger and
larger risks to save money — risks to which they had
methodically blinded themselves. The Titanic disaster suddenly
ripped away the blindfolds and changed dozens of attitudes,
practices, and standards almost literally overnight.
From PSM to OE. Ian Sutton. 2/28/08. 29
Identification of the Flags: Before
From PSM to OE. Ian Sutton. 2/28/08. 30
After
From PSM to OE. Ian Sutton. 2/28/08. 31
Seven Warning Flags
1. Stretch goals
2. "Do more with less"
3. Belief that “It Cannot Happen Here”
4. Excessive belief in rule compliance
5. Ineffective information flow
6. Ineffective audit process
7. Ignored warning signs
From PSM to OE. Ian Sutton. 2/28/08. 32
Flag One – Stretch Goals
Production creep
Production records
Initiative overload
Deferred inspection and maintenance
From PSM to OE. Ian Sutton. 2/28/08. 33
Flag Two – "Do More with Less"
Reduction of ‘Non-Essentials’
Reductions in the work force
The ‘big crew change’ (between 1983 and
2002, the number of U.S. petroleum
engineers declined from 33,000 to 18,000)
Insufficient time for detailed work
Project cutbacks
Reduced technical expertise
– Failure to follow engineering standards
– Loss of the technical backbone
From PSM to OE. Ian Sutton. 2/28/08. 34
Flag Three – Belief that “It Cannot Happen Here”
Lack of direct experience
Good occupational safety performance
Lack of imaginative thinking
From PSM to OE. Ian Sutton. 2/28/08. 35
Flag Four – Excessive Belief in Rule Compliance
Rules cannot predict the
unpredictable
Rules poor on human error
Regulators don’t understand the
industries that they regulate
Can hold back initiative (e.g., Piper
Alpha)
From PSM to OE. Ian Sutton. 2/28/08. 36
Flag Five – Ineffective Information Flow
Critical safety information is buried,
lost or diluted
Team player culture
Fear of litigation
Failure to transmit corporate memory
Mergers and acquisitions
“. . . one of the recurrent findings in disaster research is that information that
something was wrong was available somewhere within the organization but was
not communicated to the relevant decision-makers. For a variety of reasons the
bad news never landed on the desk of someone who had the authority and
inclination to do something about it.”
From PSM to OE. Ian Sutton. 2/28/08. 37
Flag Six – Ineffective Audit Process
“Softened” news to senior managers
Their baby is ugly
Failure to identify root causes
– There is no such thing as true ‘root cause’
– However, following the chain backwards will
break away from the obvious
Inadequate follow-up
Failure to follow standards
From PSM to OE. Ian Sutton. 2/28/08. 38
Softening of Bad News
Maintenance Mechanic:
“That safety valve is totally unreliable. One day its failure is going to lead to a major
explosion. We need to fix it now!”
Maintenance Supervisor:
“The safety valve is unreliable and it is in a critical service ⎯ we must repair it as soon
as possible.”
Maintenance Manager:
“The valve is important, and we will make sure it is repaired at or before the next
turnaround.”
Plant Manager:
“All valves requiring maintenance work have been identified, and will be repaired on
schedule.”
Operations Vice President:
“I am proud to report that we have an effective and proactive maintenance program —
all opportunities for improvement have been identified.”
From PSM to OE. Ian Sutton. 2/28/08. 39
Flag Seven – Ignored Warning Signs
Failure to learn from near misses
Failure to draw on experience
elsewhere
From PSM to OE. Ian Sutton. 2/28/08. 40
Section 5 – Incident Analysis
Les gens heureux n’ont pas d’histoire.
(Happy people don’t make history).
From PSM to OE. Ian Sutton. 2/28/08. 41
Root Cause Analysis
From PSM to OE. Ian Sutton. 2/28/08. 42
Categorization
Equipment failure
Human error
– Fixation
– Slips
– Mistakes
– Violations
Systems failure
But categories don't always apply
Events have causes which themselves are events, and so
on
From PSM to OE. Ian Sutton. 2/28/08. 43
Three Levels of Root Cause
Field level
– Immediate corrective actions
– Stops same event occurring at the same site and at sister facilities
Management level
– Time frame say three months
– Stops similar events
– Addresses management issues
Senior management level
– Program deficiencies
– Emphasis on goals and budgets
From PSM to OE. Ian Sutton. 2/28/08. 44
Example – Block Valve Installed During
Operation
Level One (immediate / field)
– Check other block valves are open
– Check tags are in place
– Provide a ten minute briefing to the crews
Level Two (management)
– Conduct a hazards analysis to evaluate removal of
valves
– Ensure that procedures and training policies have
been followed
– Management of change systems
Level Three (senior management)
– Evaluate owner / contractor / sub-contractor
interface
– Look at safety culture issues
From PSM to OE. Ian Sutton. 2/28/08. 45
Systemic Root Causes
Interface and contractor
management
Procurement
Failure to understand risk
From PSM to OE. Ian Sutton. 2/28/08. 46
Section 6 – System Availability and
Troubleshooting
From PSM to OE. Ian Sutton. 2/28/08. 47
Features of Availability Work
There is an objective target
It is possible to be "too good" –
unlike safety
Not looking for perfection
Quantification is key
From PSM to OE. Ian Sutton. 2/28/08. 48
Six-Step Program
From PSM to OE. Ian Sutton. 2/28/08. 49
The Opportunity
From PSM to OE. Ian Sutton. 2/28/08. 50
Troubleshooting – Where the Money Lies
From PSM to OE. Ian Sutton. 2/28/08. 51
Troubleshooting
Need understanding (education in
addition to training)
Consider simulators / emulators
From PSM to OE. Ian Sutton. 2/28/08. 52
Section 7 – Creating Operational Excellence
From PSM to OE. Ian Sutton. 2/28/08. 53
Common Features
1. Internalize risk perception
2. Understand engineering standards
3. Learn from audits and investigations
4. Restructure hazards analyses
5. Learn from events elsewhere
From PSM to OE. Ian Sutton. 2/28/08. 54
1. Internalize Risk Perception
Recognize passive hazards (blocked-in relief
valves)
Understand complexity
Education / Training
– Train in how to do something
– Educate in the why of something
Preoccupation with failure
– Relates to internalization of risk
– Relates to willingness to report and follow up
on near miss events
From PSM to OE. Ian Sutton. 2/28/08. 55
2. Understand Engineering Standards
Codes, standards, company standards,
industry good practice
Natural area for expert systems
From PSM to OE. Ian Sutton. 2/28/08. 56
3. Learn from Audits and Investigations
Relates to Warning Flags
Willingness to listen to
findings
From PSM to OE. Ian Sutton. 2/28/08. 57
4. Restructure Hazards Analyses
Complexity and weirdness of real systems
– Multiple contingencies
– Very sophisticated instrumentation
– Unidentified common cause effects
– Subtle interactions
Lack of knowledge of safe operating limits
Lack of quantification
Confusion with design reviews
For every complex problem
False confidence
there is an answer that is
clear, simple — and wrong.
Interfaces
Human error
H.L. Mencken (1880 – 1956)
Order: hazard / consequence / likelihood
From PSM to OE. Ian Sutton. 2/28/08. 58
Think the Unthinkable
Experience can be a drawback
Requires an effective hazards
analysis leader
Imaginary gardens with real
toads in them.
From PSM to OE. Ian Sutton. 2/28/08. 59
5. Learn from Actual Events
Catastrophic incidents are rare at
any one site
Agency web sites
Incident data bases
– National Response Center
– Accidental Release Information
Program
– Census of Fatal Occupational Injuries
– Major Accident Reporting System
– Marsh & McLennan Reviews
– Process Safety Beacon
Consider internal DVDs of events at
the facility
From PSM to OE. Ian Sutton. 2/28/08. 60
Conclusions
The transition from PSM to OE will
require that all employees, contract
workers and managers must understand
the systems for which they are
responsible.
They require education in addition to
training.
By using existing management tools
along with their understanding, safety,
environmental performance and
profitability will all improve.
From PSM to OE. Ian Sutton. 2/28/08. 61
