OTC 11000

Risk Assessment of a Tanker Based Floating Production Storage and Offloading

(FPSO) System in Deepwater Gulf of Mexico
Demir I. Karsan, Rajiv K. Aggarwal, Bechtel Offshore; Jon D. Nesje, Safetec; Subir Bhattacharjee, Mobil Corporation; Cyril
E. Arney, Marathon Oil; Bill M. Haire, MODEC USA; Jorge E. Ballesio, ABS

Copyright 1999, Offshore Technology Conference

extreme environments, and have different configurations and
This paper was prepared for presentation at the 1999 Offshore Technology Conference held in designs. Most are conversions from single hull tankers.
Houston, Texas, 3–6 May 1999.
Majority in the North Sea is new-built with double hulls [1].
This paper was selected for presentation by the OTC Program Committee following review of The recent installations indicate that the industry confidence in
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Offshore Technology Conference and are subject to producing from FPSOs is on the rise, and they are now being
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Offshore Technology Conference or its officers. Electronic reproduction, installed in harsher environments and deeper waters, as
distribution, or storage of any part of this paper for commercial purposes without the written
consent of the Offshore Technology Conference is prohibited. Permission to reproduce in print
permanent and larger production units.
is restricted to an abstract of not more than 300 words; illustrations may not be copied. The Figure 1 identifies the areal extent of existing platforms
abstract must contain conspicuous acknowledgment of where and by whom the paper was
presented. and the pipeline infrastructure in the US GOM and the current
deepwater offshore leases. There are over 4,000 offshore
platforms and about 20,000 miles of pipelines of various sizes
Abstract
installed in the US GOM. The majority of the GOM
Floating Production, Storage and Offloading (FPSO) systems
production facilities are supported by steel jacket platforms,
are gaining significant consideration for deepwater oilfield
tripods, or caissons in water depths up to 1,000 ft. There are a
development in the Gulf of Mexico (GOM). The risks
few steel jacket platforms in water depths exceeding 1,000ft.
associated with a FPSO system differ from those for the
MMS defines the deepwater zone as tracts with water
existing systems such as the conventional steel jacket,
depths exceeding 1,000 ft. Approximately 10 production
compliant tower, TLP, and SPAR.
facilities using novel technology systems such as the
The first phase of a Joint Industry Project (JIP) focused on
guyed/compliant tower, tension leg platform, and SPAR have
the evaluation of risks associated with the production
been installed in water depths up to 4,000 ft. All of the
operations from a GOM FPSO. The JIP objectives included
existing platforms utilize pipelines, which transport the
demonstration of the acceptability of a FPSO in the GOM;
produced oil and gas to an existing pipeline network or
identification of accidental events and FPSO components with
directly to shore facilities. Existing GOM pipeline and
high environmental pollution, loss of life, and financial risks;
platform infrastructure covers up to a 130-mile wide strip
and recommendation of reasonably practicable risk reducing
reaching to the 650-ft. water depth contour from North Padre
measures. The JIP was sponsored by 10 oil companies, 3
Island to Viosca Knoll. In some areas, this network extends to
FPSO contractors, 4 certification agencies, and MMS. The risk
nearby installations in water depths deeper than 1,000 ft. All
assessment work was done on a converted tanker FPSO
current deep water platforms are located reasonably close to
concept developed for the GOM by a major oil company.
an existing pipeline network and do not require offshore
This paper focuses on the key safety issues that are
storage and/or shuttle tanker transportation of produced oil.
important for the development of FPSOs in the GOM. The
MMS have recently awarded a significant number of deep
hazards evaluated and various risk assessment tasks
and ultra-deepwater leases that extend to 10,000-ft water
undertaken in the JIP are briefly discussed and important
depths [Figure 1]. Existing pipeline network is generally far
issues are highlighted. Additionally, the potential differences
away from most of these recent leases. Seabed topography is
in risks from a system using a new-built vessel are identified.
also more complex than what is encountered in the currently
The JIP has helped to better understand the safety of
developed areas. In some western regions of the GOM, the
FPSOs in the GOM by determining the potential
fields in 4,000 ft to 10,000 ft water depths will be as much as
improvements to layout, design, operational, and mitigation
150 miles away from the existing pipeline infrastructure. Due
measures.
to the long distances involved and technical difficulties
inherent to installing and maintaining large diameter pipelines
Introduction
in these water depths, offshore storage and transportation
Currently, there are at least 45 FPSOs installed worldwide but
using shuttle tankers becomes an economically viable and
none in the US GOM. These are operating in both benign and
2 KARSAN, AGGARWAL, NESJE, BHATTACHARJEE, ARNEY, HAIRE, BALLESIO OTC11000

advantageous option. This is one of the major drivers of the companies, 3 FPSO suppliers, 4 certifying agencies, MMS,
significant industry interest in the GOM FPSOs. and USCG (observer). The JIP brought the contractors’ and
FPSOs have a number of key differences compared to the participants’ worldwide experiences together to study and
systems that have been installed in the deepwater GOM. These make decisions on the feasibility of FPSOs in deepwater
differences are related to the FPSO ability to store significant GOM. Nine participants joined as partially contributing
volume of crude oil and offload to shuttle tankers. It is participants and made significant contributions by performing
important to identify and understand the risks associated with specific tasks and contributing data. These were used in
operations using an FPSO, and identify the areas that require Quantitative Risk Assessments (QRA) and quantifying the
attention and improvement before they are deployed in GOM. improvements made to the overall risk picture.
The regulatory regime in the GOM does not require formal Phase I JIP started with the following objectives:
risk assessment as required in UK and Norway. The US GOM • Demonstrate acceptability of FPSO risks in GOM.
operations since 1947 have resulted in good safety statistics • Identify accidental events and FPSO components with
due to the use of proven technology and operating procedures; high loss of life, pollution, and financial loss
lighter and open topside facilities; relatively benign consequences.
environment; and an early warning system allowing shutting- • Recommend reasonably practicable (cost effective)
off the platform operations and evacuation of personnel before risk reducing measures. Identify safety issues that can
extreme hurricanes. Production from deepwater in remote sites be used by participants in developing their GOM
requires use of novel production systems and technologies. FPSO designs.
Risks presented by these systems are different and require • Identify safety issues that can be used by participants
assessment. in developing their GOM FPSO designs.
A few offshore operators have used the risk and reliability • Recommend work plan (Phase II) for improving the
assessment technologies in the GOM in limited ways. Past reliability of high risk components.
GOM risk assessment applications have been focussed on The organization of the Phase I JIP work scope is shown in
specific studies leading to the development of recommended Figure 2. The Phase I work started with the contribution from
practice, evaluation of specific components of new systems, a participating major oil company, the conceptual design
reassessment of existing platforms, and providing support for details of their in-house FPSO system and its Preliminary
cost-benefit based decisions and selections. Significant Hazard Assessment (PHA), which was prepared by Bechtel
research programs have been initiated at several US and Safetec. The potential risks identified in the PHA were
universities [2]. further reviewed with the JIP Participants. QRAs were
There is significant offshore industry concern that some of performed for all major risk contributing events as identified
the proposed novel deepwater production systems are being in the PHA. These included fire and explosion events
implemented within a very short concept to field application associated with leaks from process systems, risers and
time span of two to three years, with no past industry turret/swivel arrangement. It also included events associated
experience in the GOM. Concern is also expressed on the lack with cargo oil storage and offloading to shuttle tanker, as well
of FPSO related GOM databases and the lack of application of as potential collision from vessels.
the MCAPS [3] type reliability based safety or risk assessment Additional risk assessment studies have been performed to
applications in selecting and designing such deepwater evaluate risks associated with the escape, evacuation, and
production systems. The government agencies and operators rescue operations and to compare pollution risks from shuttle
need an assurance that the risks associated with these novel tanker transportation of oil against the pipeline transportation
deepwater systems are not higher than the comparable option. Sensitivity studies were also performed to assess the
production systems in shallower waters [4]. impact of double hull, use of thrusters, moving location of the
A number of significant efforts by the US based oil Master Control Centre (MCC), and manning the vessel during
companies and other organizations are underway to evaluate a hurricane.
the feasibility of the FPSO systems in GOM. FPSOs have a Various risk reducing measures have been identified and
number of components and features for which no GOM evaluated in a coarse quantitative manner to select a few for
experience exists. For example, current GOM operational more detailed investigation in Phase II. Detailed analytical
philosophy require evacuation of all personnel onboard before studies to improve risk estimates were not performed in Phase
a hurricane, but the potential practice of abandoning a fully I. The risk picture for Phase I will include the effect of some
operational ship in the path of a hurricane may be the cause of risk reducing measures.
concern. The primary emphasis in Phase II will be on the detailed
This JIP focus is on the risk assessment of a FPSO and it is evaluation of high risk events and identification of cost
the first detailed voluntary industry-wide application of risk effective risk reducing measures proposed in Phase I.
analysis technology to a FPSO system in US waters. Development of risk based requirements for a permanently
manned FPSO, detailed evaluation of pollution risk from
Overview of the JIP Program shuttle tanker vs. pipeline transportation, provision of the
The two phased JIP started in late 1997. Phase I JIP is necessary support to MMS’ Environmental Impact Statement
sponsored by 18 participants which included 10 major oil
OTC 11000 RISK ASSESSMENT OF A TANKER-BASED FPSO SYSTEM IN DEEPWATER GULF OF MEXICO 3

(EIS) to facilitate approval of FPSOs and shuttling operations and transportation vessels (such as the helicopter). This will be
in the US GOM will be other objectives. needed to compare the overall (total) risk associated with
production from FPSOs against other production systems.

Details of Base Case FPSO and Risk Evaluation Deck Layout. The Base Case risk assessment was performed
Scope for a converted tanker with water filled wing tanks, i.e.,
The Base Case FPSO system evaluated in this JIP is shown in central storage tanks, original pumps, engine rooms, and
Figure 3. Due to the conceptual nature of the design, only accommodations at stern are retained. Thus, the flare tower is
limited system information was available. Additional details located at the bow, and turret is near bow, as shown in Figure
for sub-systems and operations were assumed as work 4. The helideck will be located near the living quarters. The
progressed, based on the experience base within the JIP offloading operations will be undertaken from the stern. The
project team and participants. The subsea wells will be located process facilities are of open type design with good
at a 3 km distance from the FPSO vessel to allow drilling and ventilation. Solid deck is provided only under vessels
well completion operations by a separate rig, while processing containing oil/condensate. The process equipment will be
and production operations will be on the FPSO. The FPSO is installed at a raised elevation, 3 m above deck. The process
assumed to be at a location 150 nautical miles (NM) from the facilities and living quarters are separated by more than 100-ft.
nearest shoreline. This location is at approximately 30 NM
from the shipping fairways, and the closest shipping route is 2 Mooring/Riser System. In the Base Case risk assessment, an
miles away. eight leg permanent catenary mooring system has been used.
The key details of this system are identified as follows: The mooring lines are a combination of spiral strand wire
• 4,000 ft water depth, rope, chain, spring buoy, and will be anchored to the seabed
• 80,000 bopd crude production, using suction pile anchors. Two 10-inch production risers, one
• Converted vessel, 12-inch gas export riser, one 6-inch gas lift riser, and one
• Permanently turret moored system, control umbilical (of non-bonded composite high-pressure
• Oil exported by shuttle tankers and associated gas by a flexible pipe) are considered. The flexible risers will be
subsea pipeline, suspended underneath the turret, through riser guides within
• Manning level – 35 persons on board, the turret cylinder, and connected at riser termination deck on
• Perform risk assessment for year-3 with 4 wells the turret. The mooring and riser spread allows for appropriate
producing, and 5 wells drilled; when the annual risk is orientation to exclude collisions. The risers are not planned to
expected to be highest, be disconnected.
• Wing tanks ballasted with water,
Escape and Evacuation Facilities. The primary escape way
• Personnel evacuated prior to hurricane.
is located on the pipe rack along middle of the process deck.
The following exclusions were made to the base case risk
The escape ways on both sides of the tank deck are considered
analysis:
as secondary as shown in Figure 5. No enclosed routes are
• No Gas lift operation considered. Mustering points are provided at both bow and
• No thrusters and DP system stern. The evacuation means considered are helicopters,
• No tug assistance during offloading operations lifeboats, and life rafts.
• No standby vessel permanently at location
• No ROV located on FPSO Risk Assessment Approach
The effects of some of the above excluded options were Figure 6 presents the overall risk assessment methodology.
evaluated in the sensitivity risk analysis task and some of This methodology is discussed in more detail and additional
these options were evaluated for potential risk reduction. background is provided in a companion paper [5]. Risk
assessment involves three activities: hazard identification, risk
Event/Consequence Envelope. Phase I Base Case risk analysis, and assessment of acceptability. The risk is estimated
analysis included the effects from all hazardous scenarios on by combining the probability of an event with the
the FPSO vessel (environmental events, drill rig operations, consequences of its escalation to personnel, pollution, and
shuttle tankers, helicopter transport, process equipment, risers, economics of the production from a system.
subsea wells, flowlines, etc.). There are no risk acceptance criteria available for the
Only the consequences on the FPSO system, including its GOM operations and none was developed in the JIP. Thus, the
turret, risers and mooring, were determined. Consequences on evaluation of risk acceptability was excluded from the project
other field systems such as the drilling, helicopter, supply work scope. The work focussed on risk analysis and
vessels, shuttle tankers, ROVs etc. were not included. No identification of measures to reduce risks associated with high
consequence evaluations were done for the drilling rig and risk components, escalation scenarios with high probabilities
shuttle tankers themselves. [See Figure 3] and consequences. A detailed evaluation of merits and
The Phase II JIP proposes to extend the consequence economics of selected risk reducing measures will be taken up
envelope to include the risks to the shuttle tanker, drilling rig, in the next phase of the project.
4 KARSAN, AGGARWAL, NESJE, BHATTACHARJEE, ARNEY, HAIRE, BALLESIO OTC11000

The following consequences were evaluated for escalation were reviewed and the required safety related modifications
scenarios from each hazardous event: were identified. Eventually, a total of 10 events/hazards were
• Fatalities (Potential Loss of Life - PLL) identified to have potential for resulting in “critical” risks to at
• Impairment of safety functions: least one consequence class. These were selected for further
• Escape Routes (ER) evaluation by QRA. PHA was done using risk matrix
• Temporary Refuge (TR) approach as shown in Figure 7 [5].
• Evacuation Systems (ES) Blowout from subsea wellheads during drilling, wireline,
• Main Support structure (MS) workover and production activities were not considered since
• Pollution – oil spill amount these have potential for major consequences to the drilling rig
only.
• Production loss/delay
The risk estimate can be done qualitatively by using the
Quantitative Risk Assessment
PHA approach and quantitatively by using the QRA
The following events were taken up for detailed QRA
technique. PHA was done for this Base Case FPSO prior to
evaluation based upon the PHA and JIP Participants’
the JIP. Thus the focus of the JIP work was on performing
recommendations:
QRA for selected events/hazards. The QRA work involved the
1. Process leaks leading to fire and explosion events
following major steps:
2. Leaks from risers and flowlines
• Identification of undesirable events and evaluation of
3. Swivel leaks
their possible escalation.
4. Failure of the mooring system
• Analysis of possible causes and frequency of 5. Cargo oil tank incidents
occurrence of the undesirable events. 6. Cargo pump room incidents
• Analysis of the likely consequences of the different 7. Engine room incidents associated with steam boiler
event sequences. 8. Collisions from merchant and other large vessels
For most of the QRAs the event tree analysis approach was passing the area
used to present potential escalation of initiating events to 9. Collisions from supply vessels and shuttle tankers
hazardous scenarios, and to determine escalation probabilities, 10. Offloading operations
consequences, and risk to personnel, environment and The pollution consequence evaluation was done for several
production [5]. The critical initiating events were identified ranges (non-critical to catastrophic) of potential pollution that
through the PHA work. The subsequent events were then were related to minor leaks, leaks from risers, leaks from
identified based upon the detailed evaluation of the Base Case rupture of one cargo tank, and leaks from more than one cargo
FPSO layout, characteristics, and operations plans. Fault trees tank. The production loss/delay consequence evaluation was
were developed for certain nodes using CARA and other also done for several ranges (non-critical to catastrophic) that
software to estimate nodal probabilities. Other nodal were related to the number of days the FPSO will not be
probabilities were determined based on databases, simple operational. The definition of these classifications is given in
models, experience or approximate reliability analysis. the companion paper [5].
Detailed consequence analyses were then performed for The details of QRA techniques, databases and software
the terminating events to determine the consequences on used are given in the two companion papers. Detailed
impairment of safety functions, pollution, and production discussion of process and riser leaks QRAs and ship system
loss/delay. Fatality analyses was also done. The consequence QRA is given in OTC 10998 [5], and detailed discussion of
of terminal events on personnel evacuation was undertaken collision events from passing vessels, and with shuttle tankers
through a separate Escape and Evacuation analysis. during offloading is given in OTC 10999 [6]. At the time of
Various databases (UK HSE, E&P Forum, WOAD, writing of this paper not all QRA results were available.
OREDA, Safetec’s TANKEX, and PARLOC) in addition to Therefore, in this paper the important issues and contributions
the Gulf of Mexico specific databases and information of various initiating events and their escalation to hazardous
provided by MMS and USCG were used. Individual company events, and the consequences to the measures assessed are
databases provided by JIP participants were also utilized, discussed. An overview of the trend of results coming out
where available. Consequence analyses were done using the from the QRAs being performed is given.
FIREX software for fire load calculations, OHRAT models for Process leaks. Damage to FPSO due to explosions will be
release calculations and COLLIDE model for the collision risk limited due to the process system design being open-type and
analysis. the equipment being raised to an elevation 3 m above tank-
deck. The process facilities and living quarters are separated
Preliminary Hazard Assessment by more than 100-ft. The gas compression equipment, fuel gas
Twenty-nine hazardous events were identified based on the system, and gas dehydration and separator skids were
FPSO system features and operational plans. These were determined to be major leak sources, which when ignited
evaluated for potential consequences to loss of life, could lead to escalation and impair the safety causing
environmental pollution, and production loss/delay. Through personnel loss, and production loss/delay.
several work group meetings, the safety systems available The escalation scenario was found to be highly critical to
OTC 11000 RISK ASSESSMENT OF A TANKER-BASED FPSO SYSTEM IN DEEPWATER GULF OF MEXICO 5

the impairment of Escape Routes (ER) and to the impairment identified. Detailed development of mooring system design
of Temporary Refuge (TR). The MCC (considered as part of criteria for manning during a hurricane stage will be
TR) located adjacent to the utilities and at more than 100 ft. undertaken in the next phase of the JIP.
from the living quarters, may be impaired by smoke ingress Ship Systems. Incidents in cargo tanks, pump room, and
resulting from pool fire at the tank deck, or by direct impact of engine room were evaluated in detail. Safetec’s TankEx
MCC from strong explosions. A pool fire at the tank-deck will database of incidents on tankers was used as basis for
make the area impassable in the bow-stern direction, thus establishing occurrence frequencies. A session was held with
impairing the Escape Routes (ER). some of the JIP participants to establish adjustment factors to
The majority of process events lead to unignited leaks that the historical data based on differences in design and operation
will have insignificant pollution and is controlled by provision of the FPSO versus the tankers being the source of the
of drainage system, coaming, shutdown and isolation system. historical frequencies.
The ignited leaks and explosion incidents escalating to cargo The main consequences of a cargo tank explosion will be
tanks could lead to impairment of cargo tanks with significant structural damage caused by the explosion pressure. A cargo
pollution, or could also threaten the structural integrity of the tank explosion would also cause damage to the process plant,
vessel. which is installed at 3 m above the tank deck. This is likely to
The potential for risk reduction by moving MCC to the result in hydrocarbon leaks followed by fire event in the
living quarters, and for providing thrusters to turn the FPSO process area. The wing tanks on the FPSO are used for water
vessel is being evaluated. The provision of additional gas ballast and as such provide a buffer towards the cargo tanks. It
detectors will also reduce the risk. An additional task is therefore considered that major pollution may only be
undertaken within this JIP is evaluating the benefits from caused by large escalating cargo tank fires, which may
protected escapeways to personnel escape. eventually lead to loss of the entire FPSO. The fatalities will
Riser/flowline leaks. The leaks from production and gas depend upon the explosion intensity and the number of
export risers within turret and sub-sea leaks in four zones were persons in the affected area.
considered. Leaks from the production risers could cause Strict adherence to procedures and measures to control the
significant pollution, depending on the time it takes to isolate incidents are very important and have generally low cost
the leak. The risk to FPSO from sub-sea gas leaks (gas export associated. Improved fire/blast protection of living quarters
riser) will depend upon the potential for gas with flammable building will reduce potential impairment of TR and safety of
concentration reaching the surface close to the FPSO, which personnel. Installation of thrusters would allow turning of the
could cause impairment of escape routes if ignited. The vessel, to reduce potential for impairment of TR and lifeboats.
potential escalation of fires within turret to the cargo tanks Collision incidents. The potential for collision of an
could cause significant pollution. Risk could be reduced by approaching passing vessel with the FPSO was evaluated.
provision of a non-return valve on the gas export riser. Limited shipping traffic data for the area considered was
Mooring failure. A preliminary mooring system design available. The data used for the analysis was developed based
with 8 mooring lines was considered in the Base Case. on the information obtained from the USCG databases. The
The majority of mooring failures are reported for Mobile wing tanks of the FPSO vessel ballasted with water would
Drilling Units (MODUs), which are designed for operating provide dual barrier against puncturing of the central cargo
and metocean conditions different than FPSOs. A simplified tanks for this system. Thus, only very large tankers at high
reliability analysis approach was used to determine speed would puncture the central tanks, which may lead to
probabilities of failure of a single mooring line followed by significant pollution and production loss.
the system failure. The potential for multiple mooring and Several measures are available for collision avoidance, and
risers failures and consequences from drifting of vessel were consequence control. These may include radar surveillance,
evaluated. The analysis indicated that in the case of 8-leg having a standby vessel guarding the FPSO, developing
mooring system, the first mooring line may fail when hazard management plans, and installation of thrusters to
subjected to metocean loads equivalent to 1000 year return reduce target for a drifting vessel on a collision course.
period, and that the mooring system may not have the ability A detailed description of collision risk evaluation, models
to sustain this load level upon failure of the first line. used, and description of event trees and probabilities and
The probability of a drifting FPSO hitting a drilling rig in consequence evaluations are given in a companion paper OTC
the area or reaching shore were found to be very low. More 10999 [6].
detailed mooring line/system analysis is necessary to better Offloading incidents. The phase of shuttle tanker
understand the system response and improve risk estimates approaching FPSO is considered high-risk. The main hazards
established in this phase by simplified approach. associated with offloading activities were identified as
In the Base Case, no thrusters were considered and the collision with FPSO and oil spillage from hose rupture. These
FPSO is assumed to be unmanned during a hurricane. In two two potential hazardous situations were analyzed.
sensitivity analyses tasks undertaken (FPSO provided with The floating hose is designed for a maximum flow rate of
thrusters, and FPSO manned during hurricane), the potential 50,000 barrels per hour per hose string. Loading hose rupture
reduction in the failure probability of the mooring system, and can occur during connection of transfer hose and offloading
change in escalation of events and consequences will be phases. The failures may occur due to structural defects,
6 KARSAN, AGGARWAL, NESJE, BHATTACHARJEE, ARNEY, HAIRE, BALLESIO OTC11000

excessive tension, excessive pressure, or extreme fishtailing. shuttling of crude from production platforms by tankers is new
The most critical events are linked to extreme fishtailing, in to the GOM operations and the risks associated will differ
case of simultaneous calm weather, wind and current in cross from the conventional case of crude transportation by
directions, and when corrective actions fail. The other pipelines. There is significant experience with pipeline
important reasons are events related to excessive tension, due transportation and MMS have maintained databases of
to breakout and no disconnection in time. pipeline failures over the years. As the industry is moving in
The control of consequences will depend upon the ability deeper water depths far away from shore where pipeline
to detect the failure and to activate the shutdown system. infrastructure does not exist, crude transportation by shuttle
Improvement in personnel training and preparedness to face tankers is an option. Therefore, a comparison of risks
accidental situation is necessary. associated with these options has been undertaken to identify
Other risks. There are several other risks that were not the issues for the GOM.
evaluated in the JIP. Some of these, such as transportation by
helicopter and boats, and occupational risks are noted to be Moving the MCC Room. The MCC is located adjacent to the
very high contributors to the overall risks. The probabilities of utility area on the process deck. Due to the essential control
these hazards were not calculated for the FPSO considered in and safety functions of the MCC it has been defined as part of
the JIP, but were taken up from previous studies to obtain the TR. Thus the risk of impairment of MCC (TR) from
approximate estimates for comparison purposes. smoke or explosion incidents will reduce when it is moved
from its current location to within the living quarters. This will
Other Tasks have direct reduction in impairment risk to TR and reduce the
Additional tasks have been undertaken to evaluate the impact risk of fatalities as less persons will be present in exposed
of variations to the Base Case system. These tasks include the areas. The risk reduction is quantified.
following:
• Escape and evacuation study Including Thrusters. The QRA tasks indicated that thrusters
• Pollution risk comparison - shuttle tanker vs. pipeline could assist in reducing risks for several events. The Base
• Sensitivity risk assessment for moving MCC room to LQ Case design without thrusters leads to FPSO aligning itself in
• Sensitivity risk assessment of including thrusters the direction of dominant metocean parameters during a fire.
• Sensitivity risk assessment for manning during hurricanes The limited use of thrusters to turn the vessel would allow risk
• Sensitivity risk assessment for FPSO vessel with a double reduction by avoiding smoke from cargo tank or process fire
hull construction to engulf the accommodation, reduce impairment of the TR
and the lifeboats, and hence reduce the personnel risk. The
Escape and Evacuation. This study evaluated the potential thrusters would also help in collision avoidance, controlling
evacuation scenarios involving hydrocarbon-related incidents vessel after failure of a mooring line, and stabilizing the FPSO
from ignited process leaks, cargo tank fire/explosion incidents, during offloading operations. The change in the risk picture
and ignited leaks from risers and flowlines. The evacuation from using a thruster was evaluated to determine the potential
scenarios related to non-hydrocarbon incidents such as ship reduction in the risks associated with the Base Case.
collision and evacuation prior to hurricane warning were also
addressed. In case of lifeboat evacuation, it is assumed that the Manning During Hurricanes. The Base Case analysis
evacuation systems and support structure should be intact up assumes that the FPSO will be evacuated upon a hurricane
to 45 minutes from the onset of an incident to muster persons warning, following the conventional GOM practice. This
to safe area (TR), perform check, asses the incident, and practice will lead to several special issues for a FPSO based
loading and launching of the lifeboat. The probabilities of production scenario compared to other systems. The FPSO
failure to muster and evacuate successfully were determined. vessel with limited storage capacity will require to be shut-
It is commonly understood that protected escape routes will down when the cargo tanks are full, thus leaving a vessel with
reduce the risk. A sensitivity case is being done to determine tanks full with crude and subjected to hurricane seas. This
the potential risk reduction from provision of a protected sensitivity study identifies the important issues related to the
escape route from the forward part of the vessel to the living alternative of having permanently manned FPSO, and a coarse
quarters. estimate of the variation in the risk picture is made. In the next
phase of this JIP, a detailed evaluation to develop safety
Pollution Risk Comparison – Shuttle Tanker vs. Pipeline. requirements for a permanently manned FPSO facility will be
Transportation risk evaluation has been excluded from the risk developed.
picture being developed for the Base Case FPSO in Phase I.
The GOM shuttling experience is limited to the lightering Double Hull FPSO Vessel. Purpose built FPSOs will
operation undertaken in the designated lightering zones to normally be built as double hull. This issue gained importance
transfer crude from very large tankers to smaller tankers for in the US because of the recent OPA90 guidelines for tankers
transportation to port stations [7]. The risks associated with operating in the US waters. These guidelines call for a phase-
lightering operations in the GOM are documented. The out of single hull trading tankers and replacement by double-
hull tankers by the year 2010. In the Base Case, the wing tanks
OTC 11000 RISK ASSESSMENT OF A TANKER-BASED FPSO SYSTEM IN DEEPWATER GULF OF MEXICO 7

are ballasted with water to provide added protection. This permanently manned FPSO
sensitivity analysis identifies the areas for differences in risks 2. Detailed risk and reliability assessments of high risk
between Base Case single hull FPSO (with wing tanks events and components
ballasted with water) with a double hull FPSO. This will 3. Detailed evaluation of risk reducing measures
impact the risk results obtained for several Base Case QRAs, 4. Evaluation of risks associated with alternate designs
such as collision from passing vessels, cargo tanks and pump of FPSO components
room explosions. 5. Comparison of risks with other deepwater production
systems
Conclusions The above tasks are proposed to be undertaken in the
This JIP has undertaken the first detailed application of risk Phase II of this JIP.
analysis techniques to determine the quantitative measures of
risks involved with operations from a tanker based FPSO Acknowledgment
located in deepwater GOM. The authors wish to express their gratitude to the JIP
The system evaluated in this JIP was at a basic design participants (ABS, BV, Chevron, Conoco, DNV, IHC Caland,
level with only conceptual level details available. This JIP LRS, MMS, Marathon, Mobil, Modec, Pennzoil, Petrobras,
through various sensitivity studies and evaluation of alternate Oryx, Quantum Offshore Contractors, Statoil, Texaco,
designs or operations has provided an understanding of the Williams Energy Services), and the management of Bechtel
benefits from variations in the layout, design, operational, and Offshore and Dovre Safetec for allowing publication of this
mitigation measures. A further detailed risk analysis and cost- paper.
benefit evaluations of specific escalation of events, of various
risk control and risk reducing measures would enable References
confirmation of the benefits identified. [1] Karsan D. I., Aggarwal, R. K., Nesje, J. D., 1999,
The use of quantitative risk analysis technique provided a “Risk and Reliability Assessment of a Tanker Based
tool to identify potential escalation scenarios and probabilities Floating Production Storage and Offloading System
of terminating events, and consequences. This JIP has in the Deepwater Gulf of Mexico”, Proc. IBC
increased awareness of risk related issues that need Conference, Olympia, London-U.K.
consideration for development of deepwater fields in the Gulf [2] Reliability of Marine Structures Program, 1996,
of Mexico. Stanford University.
The QRA work identified the vulnerability of the FPSO [3] Amoco Production Company, 1990, “Methodology
systems, which must function during accidents. These will for Comparison of Alternative production Systems
provide basis for the development of specific emergency (MCAPS),” A Report to the Joint Industry Project
preparedness, resource analysis and planning measures. Participants, December 1990.
This work has also lead to the identification of exiting [4] Oynes, C. C. and Howard, D., 1998, “Regulations of
GOM databases and further need for databases to improve the FPSOs in the Gulf of Mexico OCS,” Proc. Offshore
risk assessments. Technology Conference 1998, OTC 8768.
FPSOs have some major differences compared to the [5] Nesje, J. D., Aggarwal, R. K., Petrauskas, C.,
existing systems operating in deepwater GOM. Comparison of Vinnem, J. E., Keolanui, G., Hoffman, J.,
risks with other deepwater systems would enable assessment McDonnell, R. “Risk Assessment Technology and its
of the acceptability of risks associated with an FPSO. Application to Tanker Based Floating production
This work has also shown use of risk assessment as an Storage and Offloading (FPSO) Systems,” 31st
important tool for safe and economical development of novel Offshore Technology Conference, May 1999. OTC
offshore production systems and extending the application of 10998.
these systems to deeper water depths where the industry [6] MacDonald, A., Cain, M., Aggarwal, R. K., Vivalda,
experience is limited. To obtain full benefits, risk assessment C., Lie, O.E., “Collision Risks Associated with
must be applied progressively at various stages of a project FPSOs in Deep Water Gulf of Mexico,” 31st Offshore
cycle. Recent applications have resulted in significant safety Technology Conference, May 1999, OTC 10999.
improvements and cost savings [8]. A section on risk and [7] Marine Board, 1998, “Oil Spill Risks from Tank
reliability assessment will become part of a new API Vessel Lightering,” Prepublication Draft prepared for
Recommended Practice document (API-RP2FPX) being National Research Council, National Academy Press,
prepared for the floating systems, which will cover some of Washington, DC.
the issues identified and lessons learned in this JIP. [8] Karsan, D.I., Ross, C.G., and Eck-Olsen, C., 1996,
Industry-wide efforts are needed for collection and “Risk Assessment and Safety Assurance Program for
improvement of GOM specific databases. These will reduce the Heidrun Field,” Proc. Offshore Technology
the uncertainties in risk estimates and result in increased Conference 1996, OTC 8103.
confidence in the QRA process.
Further work should consider the following tasks:
1. Development of design and operational criteria for a
8 KARSAN, AGGARWAL, NESJE, BHATTACHARJEE, ARNEY, HAIRE, BALLESIO OTC11000

Turbine INSTR. Inert Gas Compressor Separators Glycol

MCC Generators Air Generator Boilers Module and Coolers Regenerator

Existing Platforms and

Pipelines Infrastructure
Flare
Stack
200 m
depth Accommodation Shipping Oil Separators Chemical Separation Turret Firewater
Lact. Treater Injection Pumps
Deepwater Leases contour

Figure 4 – Base Case – Deck Equipment and Facilities Layout

Figure 1 – Gulf of Mexico - Existing Infrastructure and Deepwater

Leases
Living Quarters Evacuation Route Alternative Muster Point
MCC as part of TR

Documents from Risk

Risk Assessment
Assessment and
and
Base Case Evauation Route
Contributing Risk
Risk Comparison
Comparison
System & PHA
Participants Studies
Studies

Demonstration
Demonstration
Life Boat (2 no.)
QRA
QRA
Life Raft (3 no.)

Detailed
Detailed
QRA
QRA
Lessons
Lessons Figure 5 – Base Case – Escape and Evacuation Plan
Risk
Risk Advanced Learnt
Learnt
Advanced
Reduction
Reduction Analysis
Analysis
Measures
Measures Evaluation
Evaluation

R
Risk
isk A
Annalysis
alysis P
Plan
lannnin
ingg
Risk Picture
System
System D
Definition
efinition
R isk
Work Plan for Improving the A ccep tan ce
Reliability of High Risk Components C riteria H
Hazard
azard Id
Iden
entificatio
tificationn Risk
Risk
Reducing
Reducing
M
Measures
easures
Fre
Freqque
uenncy
cy C
Con
onsseq
equuen
ence
ce
A
Ana
nalysis
lysis A
Ana
nalysis
lysis
Figure 2 – Scope of Work – Phase I JIP
R
RIS
ISK
K E
EST
STIM
IMAATTIO
IONN

R
Risk
isk P
Pictu
icture
re
R
RIS
ISK
K A
ANNA
ALLYS
YSIS
IS

U na cc ep tab le
R isk E valu atio n

A cce p tab le

Figure 6 – Risk Assessment Methodology

Frequency
H
Hig
ighh

M
Mooddeera
rate
te H ig h

LLoow
w M
M ed
ediu
ium
m

N
Neegglig
ligib
ible
le LLoow
w

R
Reem
m oote
te

N
Noonn-C
-Crit M
M in
inoorr SSeevveere
re C
Critic
riticaall C
Caata
tasstro
tropphhic
ic
Figure 3 – Base Case Field Development Plan Using an FPSO HHig
ighh
rit

MM eeddiu
ium
m Consequences
Consequences
LLooww

Figure 7 –Preliminary Hazard Assessment – Risk Matrix Approach