Copyright 2007, Offshore Technology Conference This paper was prepared for presentation at the 2007 Offshore Technology Conference held in Houston, Texas, U.S.A., 30 April–3 May 2007. This paper was selected for presentation by an OTC Program Committee following review of 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 correction by the author(s). The material, as presented, does not necessarily reflect any position of the Offshore Technology Conference, its officers, or members. Papers presented at OTC are subject to publication review by Sponsor Society Committees of the Offshore Technology Conference. Electronic reproduction, 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 is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, OTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract The fleet of Total floating production units is at present highly

increasing. They range from converted tankers to new built

mega-FPSO projects as Girassol or Dalia (Angola). In order to

better assess for the safety, environmental and economic

aspects in these developments, it has been decided to develop

and use Risk Based Inspection methodologies for all systems

and sub-systems of the units. In the past decade, RBI

methodologies have been developed and applied for the Hull

part and the Process part of the FPSO’s. This paper presents

the Risk Based Inspection planning approach that has been

developed and implemented for defining the inspection plan of

Topsides Structural Components of FPSO’s.

This approach has been applied to the topsides modules of

a specific FPSO used as reference and will be applied in the

future to the full Total fleet of FPSO’s.

Due to the large variety of components types, a qualitative

risk approach has been developed and applied. In this

approach, failure modes are described and probabilities and

consequences of failure are assessed. Risk level and

corresponding inspection effort are estimated. An overall

inspection plan is then provided for all structural components

of the topside process modules: basic modules, pipe racks

modules and manifolds. Such a methodology has the benefit to

rank the components and results in the optimising of the

inspection and maintenance effort. Typical input data and

results of the application on the reference FPSO are included

in the paper.

Very few publications on RBI methodologies for Topsides

Structural components exist and have been actually applied.

The work presented in the paper allows now to provide such

methodology. Redundancy analyses are used and Risk

Analysis is performed taking into account the consequences of

failures of Topsides Structural Components in terms of

Physical injury, Environment and Asset (Loss of production,

Unavailability, Costs of repair). The development of RBI

approach for Topsides Structural Components, in addition to

specific RBI methodologies already developed for Hull and

Process, provide a full consistent approach for the

maintenance and the inspection of offshore facilities, where all

systems and sub-systems are addressed.

Introduction Total is now operating and will operate in the future almost a

dozen of assets having the function of storage, and/or

production, and/or offloading - in short FPSO. Paper OTC-

18563 (this Conference) is giving an extensive panorama of

this fleet [1].

These Floating Units can be ship-shaped or box-shaped or

any other shape such as TLPs, SPARs, SEMIs, etc. They can

be in steel or concrete and can handle various types of

hydrocarbon products (oil, condensates, gas, LPG, LNG…).

What are common to all are the topsides that are built in

separate modules which are then put together on the main

support in the construction yard. Each module is made of a

structure that supports all process equipment.

The inspection issue

In conventional offshore development, i.e. jacket based, the

inspection of topside structure is mainly driven by the painting

deterioration and the subsequent corrosion issue. In locations

where passive fire protection is present, there may be an issue

when the ageing of this material and potential water ingress

that would develop corrosion similar to corrosion under

insulation in areas where the temperature regime is favourable.

Basically, Total current inspection standards have been

developed on these grounds.

Now the modules are installed on a floating vessel,

consideration needs to be given to fatigue induced not only by

sea state at the operating location but also by the towing phase

from the construction yard. This new working regime for

topside structures has an obvious consequence on the

inspection activity that now needs to address the detection of

fatigue cracking, in addition with the conventional metal loss.

OTC -18912

RISK BASED INSPECTION APPROACH FOR TOPSIDE STRUCTURAL COMPONENTS Michel TRUCHON - TOTAL S.A., Antoine ROUHAN and Jean GOYET - Bureau Veritas

The prioritisation issue

The design of a topside framework makes it a complex

structure. In particular, they are built with a large number of

individual members and consequently an equally large number

of welded connexions that are potential sites for fatigue cracks

to initiate. As usual in such case, both feasibility on site and

business efficiency command to prioritise connexions and

identify the ones that need to be periodically inspected.

Introducing risk

On one hand, not all welded connections are mechanically

stressed in the same way and, thanks to smart software’s, it is

now an ordinary task to rank them according to this criterion.

In other words, this ranking exercise allows identifying the

connections that have the highest probability to crack.

On the other hand, not all modules have the same role in

the process of the facility. It is easy to understand that

equipment failure in a water injection module has not the same

consequence on production or safety as in a hydrocarbon

processing one.

Probability on one side, consequence on the other, we can

see from the above that we are naturally inclined to use risk as

the driving parameter to design inspection programmes for

topside structures.

Risk-Based Inspection is now a well-known method to

design inspection programmes but it has only been

standardised for pressure systems by API 581. When it comes

to such items as topside structures, there exists no such

reference. This makes the approach presented in this paper an

original attempt to develop inspection programs based on

Risk.

Introducing the case

This approach has been developed during a recent project,

deep offshore Angola (Dalia, Figure 1). The scope was to

design inspection programs for topside structures such as:

module framework, flare structure, riser supports and interface

connexions between modules and hull, the final objective

being to review and amend Total inspection standards

accordingly (Total inspection standards are given in [3]).

Figure 1: Dalia FPSO at the construction yard

Topsides structures overview

The topsides structures of Dalia are composed of a set of 12

topsides modules, 6 pipe-rack modules located amid-ship and

2 manifold modules portside of the FPSO.

These modules are supporting the process equipments.

Basically the process is composed of two trains A & B to

process oil and gas.

Methodology

The methodology is based on the risk analysis of topsides

structures. The study focuses on the risk of fatigue failure of

structural members. As all the modules are coated or protected

by Passive Fire Protection systems, it was decided not to focus

on structural risks related to corrosion.

The basic process for providing the inspection plan based

on our risk analysis includes two steps:

o Asses the risk level – related to fatigue – of each

structural member.

o Based on the risk level, determine an inspection

frequency.

The following sections present the methodology:

o Issues with fatigues analyses are presented.

o The way probabilities and consequences are

calculated is underlined.

o Risk Acceptance Criteria are explained which allows

building the inspection plan.

o Inspection frequency in relation to the risk level is

given.

Fatigue issues

Fatigue computations

One of the main inputs of the risk evaluation is the fatigue life

of structural components. For each module, fatigue is

computed by means of beam finite element models. The figure

3 gives an overview of the topsides structural model of the

Dalia FPSO. In fatigue computations, the selection of relevant

Stress Concentration Factors (SCF) is an important issue.

Structural complexity

The structural components of topside modules are linked

together with welded connections. The RBI methodology, as

applied in this paper, focuses on the fatigue degradation

phenomena. Generally, quantitative RBI (see [2]) is used for

determining inspection frequency in relation to fatigue

degradation. However, due to structural complexity of

connections and the wide variety of welded connection types,

it was found not tractable to used quantitative approach. This

latter uses probabilistic fracture mechanics models and generic

database of crack propagation. The kind of structural

complexity is illustrated on figure 2. It represents a very fine

mesh for fatigue computation of a primary structural node of a

topside module. It is clearly seen that the number of required

detailed fracture mechanics models for each of the connection

is not practicable. For this reason, a qualitative approach has

been chosen for risk level evaluation and risk based

inspection.

Fig 2: Typical primary node of a topside structure

Fig 3: Overview of topsides structural model of DALIA FPSO

Decision trees

For each fatigue failure that is selected for risk analysis, a

decision tree is built (see figure 5 below). It describes the

sequences of events that lead to major consequences in terms

of Physical Injury, Environment and Asset (P, E, A). In these

sequences, 3 main events have been identified:

o Collapse of the module. This might happen in

case of failure of a non-redundant primary

member.

o Local collapse of structure, leading to a breach in

the process flow.

o Escalation to fire/explosion in case where a

breach appears.

The decision tree is built for all nodes. It allows

determining the probabilities and the consequences of terminal

events initiated by fatigue failures.

Probabilities evaluations

The probabilities are assessed using several inputs. The

initiating event is the fatigue failure of a structural component

of the given module. So the probability of the initiating event

is computed based on the fatigue life of the structural detail.

Then all probabilities are computed using the event trees.

Some probabilities related to events such as the event of

ignition of gas cloud or ignition of oil leaks are assessed using

the process flow diagrams and the results and assumptions of

the existing Quantitative Risk Assessment (QRA) study.

In figure 5, it is seen that the first shunting branches deal

with total collapse of the module structure. The collapse

occurrence is directly linked to the redundancy of the structure

given a fatigue structural failure. So, redundancy

corresponding to each structural element is assessed by means

of “redundancy meetings” where experts in steel works assess

the local and global redundancy of each structural component

that is analyzed in risk analysis and RBI.

Consequences evaluations

The evaluation of consequences is made through the event

tree. For each terminal event of the event tree, 3 kinds of

consequences are assessed:

o Consequences in terms of Physical

injury/Personnel (P). This is achieved by taking

into account the eventual loss of personnel

onboard the module but also consequences of

personnel on other adjacent modules in case

events escalate to fire and/or explosions. Manning

allocation is taken from the Quantittaive Risk

Assessment (QRA) study which was established

for the FPSO. For module which process water,

consequences to personnel are much lower than

modules which process oil or gas.

o Consequences to Environnement (E). For each

isolatable section, the total volume of the section

is assessed and serves as a basis for computing

gas release volumes and oil spill volumes.

o Consequences to the Asset (A) or economical

consequences. For this kind of consequences, a

“consequences meeting” was held with:

- The Company,

- The people in charge of the design of the

process equipments, which have also

participated to the HAZID and HAZOP

sessions for process,

- The RBI team,

- The people in charge of inspection and

maintenance.

This meeting used mainly process flow diagrams

for assessing the impact of the failure of any

isolatable section on the production, given a

fatigue failure – in this isolatble section - lead to

impact the process for repairs. An example of

process flow diagram is given on figure 4.

Fig 4: Typical process flow diagram used for consequences assessment

Event tree

A typical event tree used in the study is presented in the figure

5 below:

Fig 5: Typical event tree used for the evaluation of probabilities of terminal event and consequences of initiating events.

As already mentioned, the event tree uses 4 specific

probabilities:

- X: Probability of module collapse given fatigue failure.

- Z: Probability of a breach in the flow line of the supported

equipment, given no collapse of module and equipment

support failure.

- T: Probability of explosion/fire given breach in the flow

line.

- U: Probability of component fatigue failure.

Risk acceptance criteria Once the risks – probabilities and consequences - are assessed,

the ranking is performed using the Risk Acceptance Criteria

matrix (see figure 6). This matrix allows distinguishing 3

different risk levels (from the red one – the level 1 - to the

green one – the level 3 - ) which are then used for inspection

frequency attribution.

There are 3 matrices like the matrix given on figure 6, one

for the risk to personnel, one for the risk to the environment

and one for the risk to the asset. This allows assessing the risk

level of each structural component for each type of risk

(Personnel, Environment and Asset).

Figure 6: Risk matrix used for risk ranking

The highest risk level – among the three – is assigned to

each structural component and is then used for computing the

inspection frequency, which is directly linked to it.

Risk matrices are 5x5 matrices where the labels of the 5

levels are as follows:

- For probabilities: remote, extremely unlikely, very

unlikely, unlikely, likely.

- For consequences: moderate, serious, major, catastrophic,

disastrous.

Definition of levels for probabilities and consequences are

classical. It has to be noted that the definition of the 5 levels

for the consequences to the asset (economical consequences)

is specific to the unit under consideration.

Inspection frequency In this project,

• NDT inspection each 5 years was decided for all Risk

Level 1 structural components

• NDT inspection each 7.5 years was decided for all

Risk Level 2 structural components

• No NDT inspection was decided for all Risk Level 3

structural components

NDT inspection includes PFP (Passive Fire Protection)

removal for inspection of welds. For all other inspection than

the previously defined fatigue NDT inspections, inspection is

performed according to Total inspection standards [3]. That

includes in particular general visual inspection of tertiary

structures at most every year and general visual inspection of

complete structure at most every two years.

RBI outcomes The Risk Based Inspection is part of a more general data

management system. The documentation includes the

following:

- Sketches and drawings of primary and secondary

structural elements of topsides modules used for the RBI

study. The aim of these is to match the computer model

failureFatigue

beamsupportequipment

collapse ofmodule

no collapseof module

noexplosion/fire

no breachin line

Event tree for risk analysis of structural failure of topside modules

Asset

Environment

Personnel

Consequences to:

Asset

Environment

Personnel

Consequences to:

Asset

Environment

Personnel

Consequences to:

Asset

Environment

Personnel

Consequences to:

Asset

Environment

Personnel

Consequences to:

breach in line

explosion/fire

beam do notsupport

equipment

pro

ba

bili

ty l

ev

els

Consequence levels

used for fatigue and the drawings, so that any structural

changes can be reflected and traced further on.

- 3D views of the module with emphasis on the location

where the fatigue detail is to be inspected. The provided

views are based on the engineering computer model.

- 2D views of the module, highlighting the location to be

inspected. These views do exactly correspond to the

design drawings provided. The aim is to provide a good

understanding of the location to inspect.

- A set of views with typical details found on the modules.

These views make the correspondence between the 3D

and 2D views provided and the drawings of details of

structural elements, with emphasis on the location to

inspect.

A typical outcome is provided on figure 7 where the 2D

and 3D views are shown. On figure 8, the detailed drawings

are shown, which allows the inspection team to understand

what specific welds/plates to inspect given the area from the

fatigue computation.

This kind of drawing is mandatory. The fatigue

computations are based on beam finite elements. Fatigue of

structural elements is computed by means of stress

concentration factors. These reflect the complexity of the local

connection, which cannot be represented with beam elements.

It is thus necessary to provide to inspection team the local

information that is not part of 3D Beam finite elements

models, so that inspection scope reflects the investigations and

findings of the fatigue and RBI analyses.

Figure 7: 2D and 3D views showing location of inspections

Figure 8: Detailed view of scope of inspection given 3D view

The RBI provides also the risk level for each relevant

component and the associated inspection frequency. Specific

datasheets are provided and include all relevant data of the

structural component:

o ID or TAG of the component

o Means of location: module, deck, etc.

o All data related to fatigue computations

o Structural class

o A flag indicating the presence or not of Passive

Fire Protection (PFP)

o Type of inspection to be performed: NDT, CVI

etc.

o Risk level, inspection frequency

o Means of access for inspection

An example of outcome from risk analysis is presented on

the figure 9 below:

Figure 9: Example of RBI outcome of the Dalia RBI Study.

MEMBER proba conseq Risk proba conseq Risk proba conseq Risk

level level level level level level level level level

M11Z-M217 4 2 2 4 3 2 4 5 1

M211-M21A 3 1 3 3 2 3 3 2 3

M19Z-M397 3 2 3 3 3 2 3 5 1

M313-M31F 3 2 3 3 3 2 3 5 1

M19C-M19E 2 2 3 2 3 3 2 5 2

M1AB-M19C 2 2 3 2 3 3 2 5 2

M11E-M113 2 3 3 2 3 3 2 5 2

M1CB-M11E 2 3 3 2 3 3 2 5 2

M11E-M113 2 3 3 2 3 3 2 5 2

M31T-M31Y 4 2 2 4 2 2 4 2 2

personnel environnement asset

Conclusions In this paper, we have shown an example of how basic

principles of RBI methodology can be used to design

inspection programmes for topside structures in a FPSO. The

approach developed here considers fatigue cracking at welded

connexions in the structure as the main damage mechanism.

Probability of detecting such cracks is assessed considering

the use of adequate NDT techniques. Consequence is

considered from the point of view of safety to personnel,

environment impact and production impact. The risk so

determined is used to derive inspection intervals.

With this approach, we have been made able to determine

not only the more critical locations where to carry out

inspection, but also to design the inspection effort as a

function of risk to the business as a whole. We believe that the

use of this method results in an adequate compromise between

the efficiency of the inspection programme and the effort that

needs to be made to implement it.

References 1. B. Lanquetin, J. Goyet and J. Estève, “Implementing

Risk Based Inspection on our F(P)SOs: From a practical

approach to the edge of R&D”, paper OTC-18563, Offshore

Technology Conference, Houston, 2007

2. J. Goyet, D. Straub and M.H. Faber, “Risk Based

inspection planning of offshore installations”, Structural

Engineering International, Volume 12, number 3, August

2002.

3. General Specification GS EXP 211 “Plant integrity –

Minimum inspection requirements”, Total Exploration &

Production, September 2003