Energy Practice

RISK ENGINEERING POSITION PAPER – 04

PROCESS SAFETY PERFORMANCE

INDICATORS – PSPIS
ii • Process Safety Performance Indicators (PSPIs)
(PSPIs)
 RISK ENGINEERING POSITION PAPER

CONTENTS
Section Title............................................................................................................................................... Page

1. Background.........................................................................................................................................1

2. Objective.............................................................................................................................................2

3. Scope...................................................................................................................................................2

4. Specific Requirements.........................................................................................................................3
4.1 Policy and procedure..........................................................................................................................3
4.2 Ownership and development.............................................................................................................3
4.3 Identification of PSPIs.........................................................................................................................4
4.4 Use of PSPIs in improving safety management..................................................................................6

5. Reference to industry losses...............................................................................................................6

6. Reference to industry standards.........................................................................................................6

7. Appendices..........................................................................................................................................8
7.1 Appendix A – Classification of Process Safety Incidents.....................................................................8
7.2 Appendix B – Classification of Loss of Process Containment (LOPC) for Major and Minor incidents. 10

Marsh • iii
1. BACKGROUND
The process industry has a long history of major incidents the physical systems; instrument testing and calibration
that are well-publicised with many valuable lessons to be systems are designed to ensure the integrity of the logical
learned. It has been recognised that the focus on personal systems; management systems such as procedures,
safety is one side of preventing accidents. However, a competency and performance management, and audits
major incident can cause as much harm to people in an are designed to ensure that all of the other systems work
instant as all the injuries that have been avoided in a effectively.
decade of personal safety programmes. The underlying
causes of major incidents are often related to failures in
As with any aspect of management systems, it is
what is commonly known as ‘process safety management’.
important to have metrics in place to understand how
effectively a system is working to deliver its objective.
The primary aim of process safety management is to The metrics for process safety management systems are
reduce the risk of a major accident by controlling major commonly called ‘process safety performance indicators’
accident hazards. A major accident is typically defined as (PSPIs).
a fire, explosion, or a significant release of
environmentally harmful or toxic material with costly or
PSPIs can be separated into leading and lagging
damaging consequences in terms of fatalities and injuries,
indicators; leading indicators being those that precede a
pollution, loss of revenue, damage to the assets, and/or
failure of the process safety management system and
company reputation.
lagging indicators being those that follow a failure of the
process safety management system. For example, a
Such consequences are generally defined in the measurement of the completion of pressure safety valve
organisation’s risk assessment matrix by class (PSV) testing would be a leading indicator and the
(human, environment, financial, reputation) and by number of times PSVs fail to lift at the set pressure (in use
degree. or during testing) would be a corresponding lagging
indicator.
The range of process safety activities is enormous –
everything from standardising routine operator rounds to PSPIs should be tailored to show how well the relevant
modelling the capacity of the flare system. Such activities barriers are being maintained (a leading indication) and
can prevent incidents if they are done well, or cause how well they are working (a lagging indication).
incidents if they are not done well, or not done at all.
These activities are the barriers that could stop an
The development of an effective PSPI system can provide
incident before it starts or control it if it does. Process
a clear view on how well process safety is being managed
safety management is concerned with putting these
at a site and across the wider organisation. Common PSPI
barriers in place and maintaining them so that they work
systems can allow comparisons to be made and lead to
effectively. These barriers can be physical systems,
more focused knowledge sharing – from proactive
instrumented systems, or management/people systems.
information as well as reactive.
They rely heavily on competence and diligence across all
disciplines and at all levels in the organisation.

All of the above barriers are interdependent. For

example, closed systems are designed to physically cope
with the operating pressure; instrumented systems are
designed to control the pressure within the design limit;
additional physical systems are installed to relieve the
pressure if the logical system fails to control it;
inspection and maintenance systems are designed to
ensure integrity of
1 • Process Safety Performance Indicators (PSPIs)
 RISK ENGINEERING POSITION PAPER

2. OBJECTIVE 3. SCOPE
The objective of this position paper is to define the The scope of this position paper includes the
standards rated by Marsh as very good for a set of identification and application of PSPIs in the oil, gas, and
process safety performance indicators in the oil, gas, and petrochemical industry.
petrochemical industry. These standards are
incorporated in the Marsh energy risk ranking criteria.
They can be used to support and define risk
improvement recommendations and also to provide
detailed advice to clients seeking to better understand
and improve their process safety performance.

Marsh • 2
4. SPECIFIC REQUIREMENTS
3 • Process Safety Performance Indicators (PSPIs)
4.1 POLICY AND PROCEDURE
PSPIs should be part of the corporate strategy to reduce
exposure to major accident hazards. As such, the site
measures should, where possible, relate to and roll up
into the corporate performance measures. In well-
developed corporate systems there are common
standards and definitions used to set and monitor site
and business sector PSPIs. Where this is the case, they
should be defined in a corporate policy and procedure
(P&P) and reflected in a site P&P as part of the health,
safety, and environment management system (HSE-MS).

The P&P should define the PSPIs so that all monitoring

and reporting is carried out on a consistent, comparable
basis.

Many sites are now basing their PSPIs – certainly their

lagging PSPIs – on API RP 754, PSPIs for refining and
petrochemical industries. It is essential that organisations
should adopt the measures that reflect their major
accident exposures and the control features they have
put in place to prevent them.

Corporate and site annual objectives and five-year

plans should include PSPI targets (current and next
year) and aspirations (five-year horizon).

4.2 OWNERSHIP AND

DEVELOPMENT
PSPIs for the site should be owned by the site leadership
team and accountability for the input should be
delegated to the various departments and plant area
teams. It is recommended that a senior management
team member acts as a ‘champion for PSPIs’ during their
development.

PSPIs should ideally be derived from a rigorous process

that identifies the key major accident hazards and uses
the accident trajectory analysis to work out the likely
causes and associated control measures. PSPIs can then
be based on causes identified in this analysis as the most
likely to occur and linked to the most serious
consequences. Some organisations use a bow-tie
analysis and layers of protection analysis (LOPA) to gain
insight into their risk exposures.
Alternatively PSPIs could be derived from a
combination of:

• Facilitated consultation with a cross section

of workforce.
• Review of the site safety case.
• Analysis of the incident and accident database.
• Learning from external incidents.

Opportunities for selecting PSPIs should consider the

risk control system and identify leading (those that are
evident before the incident occurs) and lagging
indicators (those that are evident after the incident
occurs).

It is recommended that the selected PSPIs are linked to

the barriers that are identified as being of particular
significance to the prevention and mitigation of process
safety accidents on the site.

For example, critical barriers that should be

considered relevant for process safety include:

• Plant design.
• Staff competence.
• Operational procedures.
• Permit to work.
• Communication.
• Instrumentation and alarms.
• Plant change control.
• Inspection and maintenance.
• Emergency arrangements.

For barriers identified as being of particular relevance,

consideration should be given to the development of
both leading and lagging indicators.

This should be a continuous process to sharpen the

focus and maintain the set of PSPIs up-to-date with
changes in the plant and the standards. A model set of
PSPIs would include no more than eight measures,
reported on a monthly basis to site management and
with an annual review of the measures to ensure the
effectiveness of the PSPIs.

The range of PSPIs should include high level site

measures, not necessarily limited to those reported to
the corporation, plus a subset of area or departmental
measures.
 RISK ENGINEERING POSITION PAPER

4.3 IDENTIFICATION OF PSPIS

There is a wide range of potential PSPIs. The classification of PSPIs can be divided into leading and lagging indicators
(see Appendix A).

The following is considered to be a typical practical selection for a site to use.

PSPI CATEGORY PSPIs

• Specific types of incident such as:
– Loss of containment incidents – major and minor categories; both including and excluding
flaring and relief valve activation.
INCIDENTS – Process unit trips – complete unit or section trips.
– Fires and explosions.
– Operating window excursions – operating outside limits for more than a given amount of time
in a defined period.
• Field audits (% of permits) – compared to a set target (not necessarily 100%).
• Document audits (% of permits) – compared to a set target (not necessarily 100%).
PERMITS
• Permit non-compliances (open/closed) – note the trend is more important than the absolute
number and any comparison across plants/companies would be difficult due to variable standards.
• Site leadership team plant visits (man-hours/year).
AUDITS • Plant safety audits (number/month).
• Audit action items overdue (annual number and %).
• Overdue tests (annual number and %).
SIS TESTING/COMPLIANCE • Testing failure to danger (annual number and %) – note the need to define tolerance.
(ESD/EDP/TRIPS/ROIV) • Failure on demand (events/year) – link to process unit trip measurements.
• Trip bypasses registered with durations >1 day; >1 month and >3 months.
• Overdue inspections without waiver.
• Inspection waivers (annual number and % of inspections).
INSPECTION • Overdue inspection recommendations (number and %).
• PSV pre-pop test fail to danger (number and % of tests).
• Number of leak clamps installed.
• Completion of all scheduled tests (%) (firewater pumps run/capacity tests, deluge systems
FIRE PROTECTION activation and application, F&G alarms, fire and gas detectors – it may be useful to break this down
by category).
• Call-in and pager testing (% success rate).
• Crisis management muster drills (number versus target).
EMERGENCY RESPONSE
• Crisis management exercises (number versus target).
• Plant drills (number versus target – by shift team).
• Daily/shift safety checks completion (%).
• Periodic safety checks completion (%).
OPERATIONS ROUTINES • (CSO/CSC, PSV status, drains/blanks, safety equipment, critical valves, and actuators position).
• Housekeeping audits by operators (%).
• Start-up/shutdown pro-forma log completion (%).
• Standing alarms (number per console).
ALARM MANAGEMENT • Alarm rate – normal operations (no/operator/hour).
• Alarm faults – priority 1 and 2 (number/%).
• MoC – number of open MoCs by year and number of overdue temporary MoCs.
• Scheduled hazard reviews (HazOps) completed and associated action items completed.
PROCESS SAFETY • Procedure updates (annual % of scheduled updates).
MANAGEMENT SYSTEMS • Staff safety critical competencies compliance (%).
• Contractor competency compliance (%).
• PHA programme compliance – annual % of scheduled updates completed.
Marsh • 4
 NOTES: HEIGHTENED RISK EXPOSURE
For the purpose of this position paper, the Heightened risk exposure cases should be reported and
following abbreviations apply: investigated as though they were actual incidents:
• Identify heightened risk exposure cases
SIS Safety instrumented system through monitoring, auditing and reporting.
ESD Emergency shutdown
EDP Emergency depressurisation • Encourage reporting of heightened risk exposure cases
ROIV Remotely operated isolation valve – educate personnel to recognise process safety
F&G Fire and gas hazards as well as personal safety hazards.
CSO Carseal open
CSC Carseal closed • Investigate serious heightened risk exposure cases
MOC Management of change to the required degree and determine corrective
HAZOP Hazard and operability review actions.
PHA Process hazard analysis
• Incorporate in PSPIs and analyse trends.
The P&P should define the PSPIs so that all monitoring
and reporting is carried out on a consistent comparable SUMMARY
basis. The foregoing discussion of the process safety
incident classification can be summarised as follows:
The reporting of PSPIs can be split into four categories: • Structure process safety performance indicators in
relation to the classification of process safety
MAJOR INCIDENTS incidents.
Major process safety incidents should be investigated to • Educate personnel to recognise lower order
determine corrective actions. For major incidents there incidents and situations.
will be specific reporting and investigation protocols to
be followed and these may involve the authorities and • Encourage reporting of heightened risk exposures
may require investigation by a special team. and near-miss incidents.

• Investigate significant heightened risk exposures,

MINOR INCIDENTS
near- miss incidents, and major/minor incidents to
Minor process incidents should be investigated
determine corrective actions.
to determine corrective actions. Minor process
safety incidents should be incorporated in the • Establish regular external assessments of process
site PSPIs. safety standards.

NEAR MISSES • Learn from external incidents.

Reporting of near-miss process safety incidents should • Use reported process safety incident data to identify
be encouraged across the site; repeated near-miss key areas and aspects of exposure.
incidents will eventually lead to actual consequences.
The whole workforce should be educated to recognise
near-miss incidents and report them. This requires a ‘no
blame’ culture to encourage open reporting.

Near-miss process safety incidents should be investigated

to determine corrective actions.

Near-miss process safety incidents should be

incorporated in the site PSPIs.

5 • Process Safety Performance Indicators (PSPIs)

 RISK ENGINEERING POSITION PAPER

4.4 USE OF PSPIS IN IMPROVING

SAFETY MANAGEMENT
PSPIs should be part of the site safety management
system and used specifically to monitor performance and
identify opportunities for improvement.

For example, the number of fires could be a typical

lagging PSPI. If the number was to increase or to be
significantly higher than the number for peer comparison
organisations then the range of causes of fires could be
further analysed to identify a significant common factor
or factors. These factors could then be the subject of
further PSPIs at the site. Ideally, these would be a mix of
leading and lagging PSPIs.

PSPIs should be reported along with the other key

performance indicators (KPIs) for the site and in
particular alongside the more traditional safety
performance measures such as lost time injury (LTI)
frequency rates etc. As such, they should be reviewed by
the site safety steering committee to determine follow-
up actions and to provide the basis for communication
to the site personnel.

Corporations should use PSPIs to identify best practices

and to manage risk exposures.

5. REFERENCE TO accident hazards.”

INDUSTRY
LOSSES
The report of the BP US Refineries Independent Safety
Review Panel (The Baker Panel Report) January 2007,
based on the BP Texas City refinery incident 23 March
2005:

• Recommendation #7 refers to establishing leading

and lagging performance indicators for process
safety.

The major accident investigation report, prepared by the

UK HSE on three incidents that occurred at the BP
Grangemouth complex, between 29 May 2000 and 10
June 2000:

• Recommendation #4 on Group Safety Assurance – “BP

should review its Group safety assurance process as a
key part of corporate assurance process. In particular
BP should develop performance measures for major
6. REFERENCE TO Process Safety Leading and Lagging Metrics, Center for

INDUSTRY STANDARDS Chemical Process Safety (CCPS), American Institute of

Chemical Engineers (AIChE), 2011.
HSG254 Step-By-Step Guide to Developing Process
Safety Performance Indicators, UK Health and Safety API Recommended Practice 754, Process Safety
Executive (HSE), 2006. Performance Indicators for the Refining and Petrochemical
Industries, American Petroleum Institute (API), 2010.

Marsh • 6
77 •• Process
Process Safety
Safety Performance
Performance Indicators
Indicators (PSPIs)
(PSPIs)
 RISK ENGINEERING POSITION PAPER

7. APPENDICES
APPENDIX A – CLASSIFICATION uncontrolled
OF PROCESS SAFETY INCIDENTS
Personal safety incidents are well recognised from
definitions such as fatality, lost time injury, medical
treatment, first aid, near miss, and hazard. These are
often shown as hierarchy or pyramid where the
frequency of occurrence of the lower order incidents is
said to influence the likely occurrence of the higher
order incidents as they reflect the overall safety culture
in the organisation.

Personal safety incidents have historically received more

attention and are more readily reported and investigated.
Process safety incidents and heightened risk exposures
with low or no actual consequences are not so readily
recognised or reported. In general, personal safety
incidents tend to be higher frequency and low
consequence, affecting individuals in most cases; in
comparison, process safety incidents may appear to be
less frequent, but typically have major consequences –
actually or potentially affecting larger numbers of people
as well as the environment, the community, the assets,
profitability, and reputation. Obviously, the process safety
incidents with significant actual consequences are
notified and thoroughly investigated; however, the
equivalent near-miss incident may not even be
recognised. Consequently, there may be an underlying
problem with process safety management that does not
become evident until the major incident occurs. In order
to bring the underlying process safety issues to the
attention of the management team, it is necessary to
ensure that all incidents and near-miss situations are
recognised, reported, and investigated. The following
hierarchy is intended to provide the framework for this.

A parallel process safety incident hierarchy can be

illustrated as follows by a similar pyramid of
consequences as that which is often used for personal
safety; each of the incident types is described in more
detail below.

The following definitions for ‘major’ and ‘minor’

incidents have been developed using reporting criteria
from the CCPS and API 754 Process Safety Performance
Indicators guidance documents. The reporting criteria for
loss of primary containment (LOPC) are based on the
United Nations recommendations on transportation of
dangerous goods, section 2. The objective is to provide a
recognised system for defining threshold quantities of
 MORE
LAGGING
MAJOR INCIDENT
INDICATO
RS This is an incident with major/catastrophic
consequences (equivalent to API 754 tier 1 process
safety event) defined as follows:

• Fatality or hospital admission.

MAJOR
• LOPC (see appendix B for threshold quantities).
INCIDENT
• Fire or explosion resulting in direct company loss
>US$25,000. MINOR
INCIDENT
Examples of Major Incidents:

• Vapour cloud explosion

NEAR MISS (VCE) from hydrocarbon
INCIDENT
leak causing fatalities and major damage.
• Floating roof full surface fire taking several days
to extinguish. HEIGHTENED
RISK EXPOSURE MORE LEADING
• Spill fire beneath vacuum tower causing collapse.
INDICATORS

• Ship colliding with jetty and causing major damage.

SAFETY INCIDENT PYRAMID APPLIED TO PROCESS SAFETY
• Uncontrolled internal tube leak resulting in
complete destruction of furnace.
• Total power failure causing crash shutdown extended
by consequential damage.
releases of MINOR INCIDENT Marsh • 8
material.
This is an incident with minor moderate consequences rounds and level reduced below safe filling height.
(equivalent to API 754 tier 2 process safety event) defined
as follows:
• Recordable injury.
• LOPC (see appendix B for threshold quantities).
• Fire or explosion resulting in direct company loss
>US$2,500.
Examples of minor incidents:
• Pump seal leak – hydrocarbon released but not ignited.
• Emergency shutdown of process unit due to heater
tube failure – repaired within 3 days.
• Leak on hydrocarbon piping – weld failure due
to construction defect.
• Flexible hose bursts on truck loading rack – manual
ESD activated by operator.
• Plug blows out of sight glass on fluid catalytic cracker
(FCC) fractionator – isolated and replaced by
operator.
• Floating roof rim seal fire due to lightning strike
– extinguished by foam pourers.
• Explosion within furnace on light off due to
ineffective purging by operator; minor damage to
refractory.

NEAR-MISS INCIDENT
This is an incident with no actual consequences; however,
if the circumstances were slightly different there could be
serious consequences. In many cases the first barrier fails
but subsequent barriers or fortuitous intervention
prevents the full development of the incident. A near-
miss incident is equivalent to API 754 tier 3 process safety
events.
Examples of process safety near-miss incidents:
• ESD valve fails to close automatically but operator
responds to alarm and closes the valve manually from
the local station.
• Minor leak from hydrocarbon line due to incorrect
gasket on pump discharge flange – able to switch to
spare pump, isolate and fix the fault without a plant
shutdown.
• High liquid level in flare knockout drum; level indicator
faulty – but problem identified by operator visual
checks before liquid carried over to flare stack.
• Defective construction weld on hydrocarbon
line discovered by inspection at turnaround.
• Interlock bypass used for start-up but kept on
until noticed by relieving panel operator two
days later.
• Tank filled above safe filling height without activating the
alarm but no spill occurs; picked up by operator on
 HEIGHTENED RISK EXPOSURE • Inspection waiver not risk assessed and not
sanctioned according to site policy.
A heightened risk exposure is an action or a lack of action
that increases the likelihood or consequences of a potential • MoC process failed to specify operator training
required before new system started up.
incident. A heightened risk exposure is also a significant
gap in process safety management standards compared • New plant started up with construction blinds used
to world class. for isolation.
• Furnace ESD function tested but ESD valves not tested
There is no actual incident or event in these cases. to verify tight-shutoff (TSO) capability.
Heightened risk exposures are equivalent to unsafe acts • A number of large hydrocarbon inventories are not
or unsafe conditions that are recognised as personal protected by remote operated isolation valves – this
safety hazards. Failure to perform risk control activities would be an example of a gap in risk control standards
compared to world-class practices.
as required by site/company and actions that could lead
to or increase the potential consequences of an incident Gaps versus world-class process safety management
would be regarded as heightened risk exposures. There standards present a continuous exposure. These gaps
would also be a heightened risk exposure if there are are typically identified by:
significant gaps in risk control standards compared to
world-class process safety management standards. A • Site incident investigation.
heightened risk exposure is equivalent to API 754 tier 4 • Process hazard analysis.
process safety events. • External audits.
Examples of heightened risk exposures: • Process safety reviews (internal and external).
• Work permit non-compliance: error on equipment number. • Learning from incidents (external).
• Diesel firewater pump extended outage – total
Once identified, the decision on whether to close these
reliance on electrically driven machines and therefore
exposed to power failure during fire emergency. gaps should reflect company/site policy on risk
management.

9 • Process Safety Performance Indicators (PSPIs)

 RISK ENGINEERING POSITION PAPER

APPENDIX B – CLASSIFICATION OF LOSS OF PROCESS CONTAINMENT

(LOPC) FOR MAJOR AND MINOR INCIDENTS

MAJOR INCIDENT (EQUIVALENT TO API 754 TIER 1 PROCESS SAFETY EVENT)

MATERIAL HAZARD CLASSIFICATION THRESHOLD QUANTITY THRESHOLD QUANTITY

(OUTDOOR RELEASE) 1 (INDOOR RELEASE) 1

Toxic inhalation hazard zone A materials 5 kg 2.5 kg

Toxic inhalation hazard zone B materials 25 kg 12.5 kg
Toxic inhalation hazard zone C materials 100 kg 50 kg
Toxic inhalation hazard zone D materials 200 kg 100 kg
Flammable gases 500 kg 250 kg
or
liquids with initial boiling point <35˚C and flash
point <23˚C
or
other packing group I materials (excluding strong
bases and acids)
Liquids with initial boiling point >35˚C and flash 1000 kg 500 kg
point <23˚C or Or
or 7 barrels 3.5 barrels
other packing group II materials (excluding
moderate bases and acids)
Liquids with initial boiling point >35˚C and flash 2000 kg 1000 kg
point <60˚C or Or
or 14 barrels 7 barrels
liquids with flash point >60˚C released at a
temperature at or above flash point
or
strong bases or
acids or
other packing group I materials

Notes:

1. United Nations recommendations on transportation of dangerous goods, section 2.

Marsh •
10
 MINOR INCIDENT (EQUIVALENT TO API 54 TIER 2 PROCESS SAFETY EVENT)

MATERIAL HAZARD CLASSIFICATION THRESHOLD QUANTITY THRESHOLD QUANTITY

(OUTDOOR RELEASE) 1 (INDOOR RELEASE) 1

Toxic inhalation hazard zone A materials 0.5 kg 0.25 kg

Toxic inhalation hazard zone B materials 2.5 kg 1.2 kg
Toxic inhalation hazard zone C materials 10 kg 5 kg
Toxic inhalation hazard zone D materials 20 kg 10 kg
Flammable gases 50 kg 25 kg
or
liquids with initial boiling point <35˚C and flash
point <23˚C
or
other packing group I materials (excluding strong
bases and acids)
Liquids with initial boiling point >35˚C and flash 100 kg 50 kg
point <60˚C or or
or 1 barrel 0.5 barrel
liquids with flash point >60˚C released at a
temperature at or above flash point
or
other packing group II materials (excluding
moderate bases and acids)
or
strong bases and acids
Liquids with flash point >60˚C released at a 1000 kg 500 kg
temperature below flash point or or
or 10 barrels 5 barrels
moderate bases and acid

Notes:

1. United Nations recommendations on transportation of dangerous goods, section 2.

11 • Process Safety Performance Indicators (PSPIs)

 RISK ENGINEERING POSITION PAPER

THE ENGINEERING SERVICES TEAM

Marsh’s Risk Engineering Services
permit a benchmarking system
team has been established for
which gives a global opinion of the
over 25 years and is uniquely
risk, assessed against peer plants
qualified to provide risk managers
world-wide.
and underwriters with the
essential information they need
to determine the right limit and From the earliest planning stage to
scope of cover and the right price. the last operational phase, the
engineering services team is able
to contribute practical and
Each member of the team is a
cost-effective advice, and
qualified engineer, with practical
assistance.
experience in design,
construction, operation, and
maintenance across a broad In addition to tailored
range of oil, gas, and programmes, the team has a
petrochemical risks. series of core packages, covering
everything from managing a
major emergency to risk
They have all been trained in
reduction design features, and
advanced insurance skills, in the
safe working practices.
ability to assess and analyse risk,
and to communicate effectively
and frequently in more than one The Engineering Services team
language. uses its breadth of expertise,
experience, and its practical
knowledge and skills to
The goal is to build bridges
communicate a real understanding
between risk engineering,
of physical risks, your insurance
insurance and risk management,
implications and the commercial
and between the client and the
operating environment.
underwriter. At the same time,
the comparative skills of the
team

Marsh •
12
For further information, please
contact your local Marsh office or visit
our web site at: marsh.com

BEIJING LONDON OSLO

Tel: +86 10 6533 4070 Tel: +44 (0)20 7357 1000 Tel: +47 22 01 10 00
Fax: +86 10 8529 8761 Fax: +44 (0)20 7929 2705 Fax: +47 22 01 10 90
CALGARY MADRID PERTH
Tel: +1 403 290 7900 Tel: +34 914 569 400 Tel: +61 8 9289 3888
Fax: +1 403 261 9882 Fax: +34 913 025 500 Fax: +61 8 9289 3880
CAPE TOWN MOSCOW RIO DE JANEIRO
Tel: +27 21 403 1940 Tel: +7 495 787 7070 Tel: +55 21 2141 1650
Fax: +27 21 419 3867 Fax: +7 495 787 7071 Fax: +55 21 2141 1604
DUBAI MUMBAI
Tel: +971 4 223 7700 SAN FRANCISCO
Tel: +91 226 651 2900 Tel: +1 415 743 8000
Fax: +971 4 227 2020 Fax: +91 225 651 2901 Fax: +1 415 743 8080
HOUSTON NEW YORK
Tel: +1 713 276 8000 SINGAPORE
Tel: +1 212 345 6000 Tel: +65 6327 3150
Fax: +1 713 276 8888 Fax: +1 212 345 4853 Fax: +65 6327 8845

Marsh is one of the Marsh & McLennan Companies, together with

Guy Carpenter, Mercer, and Oliver Wyman.
The information contained herein is based on sources we
believe reliable and should be understood to be general risk
management and insurance information only. The
information is not intended to be taken as advice
with respect to any individual situation and cannot
be relied upon as such.
In the United Kingdom, Marsh Ltd is
authorised and regulated by the
Financial Conduct Authority.
Copyright © 2015 Marsh Ltd
All rights reserved.