International Journal of Pressure Vessels and Piping 86 (2009) 329–334

Contents lists available at ScienceDirect

International Journal of Pressure Vessels and Piping

journal homepage: www.elsevier.com/locate/ijpvp

Assessment of local thin areas in a marine pipeline by using the FITNET FFS
corrosion module
S. Cicero a, b, *, R. Lacalle a, b, R. Cicero a, b, D. Ferreño a, b
a
Dpto. Ciencia e Ingenierı́a del Terreno y de los Materiales, Universidad de Cantabria, ETS. Ingenieros de Caminos, Av. Los Castros s/n, 39005 Santander, Cantabria, Spain
b
INESCO Ingenieros SL, Centro de Desarrollo Tecnológico de la Universidad de Cantabria, Fase A, Mod. 2003, Av. Los Castros s/n, 39005 Santander, Cantabria, Spain

a r t i c l e i n f o a b s t r a c t

Article history: This paper analyses the structural integrity of the marine stretch of a pipeline which is placed over
Received 11 December 2007 a natural bay. The pipeline is part of a 30-year-old installation used for the provision of petrochemical
Received in revised form products to a nearby chemical plant. Although there have been no relevant leaks in the past, both the
31 October 2008
visual inspections performed (revealing numerous local thin areas) and the fact that it is located in
Accepted 18 November 2008
a highly sensitive place with high ecological and tourist value recommend the assessment of the pipeline
in order to ensure that it is working in safe conditions and that there are no risks for the environment or
Keywords:
the people living in the surrounding area. The assessment has been performed using the newly devel-
Local thin areas
Structural integrity oped FITNET FFS procedure, whose local thin areas assessment methodology is also explained and
FITNET FFS compared to the analyses proposed by other well known procedures.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction Before the FITNET FFS procedure was developed there were
numerous assessment procedures of a clearly national nature
During the visual inspection of an approximately 1800 m long (British BS7910 [5] or the French RCC-MR [6]) and others whose
marine stretch of a pipeline placed over a natural bay, many local application was restricted to a specific industry (i.e, R5 [7] and R6
thin areas (LTA) were detected (Fig. 1 shows an example) in the [8] of British Energy and the Swedish SAQ [9] in the nuclear sector).
exterior of the pipe. This pipeline, whose total length is around Moreover, each of these procedures tended to focus on one specific
5000 m, is used for the provision of petrochemical products type of failure (R5 on creep, BS7910, R6 and SAQ on fracture.),
(conversion oil) to a chemical plant and its failure would inflict there being no procedure which allowed a component to be
substantial damage on the environment and the population living assessed for the various failure mechanisms.
nearby. This makes it essential to assess the pipeline in order to At the same time, some other industrial powers, such as the USA
ensure that it is working safely. and Japan had their own procedures offering a simpler, more clearly
The structural integrity analysis of the pipeline has been per- defined picture. Thus, American procedures such as API 579 [10] in
formed using the corrosion module of the FITNET FFS procedure [1– the petrol sector and ASME XI [11] in the nuclear industry or the
4] and requires the following steps: structural analysis, stress Japanese JSME procedure [12] were fully defined and widely used
analysis and assessment of the LTA (all of these are explained in industry.
below). The FITNET FFS procedure has been the main result of the This panorama in Europe, characterised by a lack of uniformity,
recently completed European Fitness for Service Network [1] and an excessive variety, lack of global vision and delay with respect to
constitutes a unified European procedure for structural integrity other countries made it absolutely essential to develop a unified
assessment, which covers the analysis of components and struc- European procedure covering the main failure modes and offering
tures under four main failure mechanisms: fracture-plastic European industry a tool with which to design, manufacture and
collapse, fatigue, creep and corrosion. manage industrial components and installations in a safer, more
efficient way.
Finally, the procedure is organised in three volumes (proce-
dure [2], annex [3] and case studies and tutorials [4]) and has
* Corresponding author. Dpto. Ciencia e Ingenierı́a del Terreno y de los Materi-
ales, Universidad de Cantabria, ETS. Ingenieros de Caminos, Av. Los Castros s/n,
been submitted to the European Committee for Standardisation
39005 Santander, Cantabria, Spain. Tel.: þ34 942 200917; fax: þ34 20 18 18. (CEN) for its adoption as a European standard for structural
E-mail address: ciceros@unican.es (S. Cicero). integrity.

0308-0161/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijpvp.2008.11.021
330 S. Cicero et al. / International Journal of Pressure Vessels and Piping 86 (2009) 329–334

Nomenclature tmC minimum allowable remaining wall thickness to

prevent circumferential failure
Do outer diameter tmL minimum allowable remaining wall thickness to
Di inner diameter prevent longitudinal failure
l length or longitudinal dimension of the thin area w width or circumferential dimension of the thin area
LTA local thin area aelbow elbow parameter quantifying the elbow effect on the
M applied bending moment hoop stress
n parameter used for the calculation of scyl belbow elbow parameter quantifying the elbow effect on the
P working pressure axial stress
Q length correction factor l parameter used to quantify belbow
R average radius in the transversal section of the elbow sa axial stress from the axial force in the bend
(as shown in Fig. 6) sA code allowable stresses
Rb average radius of the elbow (as shown in Fig. 6) scyl nominal failure stress of an unflawed pipe
RSFa allowable residual strength factor (used in API 579) sf flow stress (average value between the yield stress and
rsfECmin minimum required strength factor for circumferential the ultimate tensile strength)
failure for the elbow system stress sh hoop stress
rsfLmin minimum required remaining strength factor for the ssys system stress (sa þ sz)
nominal hoop stress suts material ultimate tensile strength
t pipe thickness syield material yield stress
tm allowable remaining wall thickness for combined sz axial bending stress
internal pressure and axial loading ðmax½tmL ; tmC Þ

Concerning the assessment of LTA, the FITNET FFS procedure analysed here. All the pipes are supported every 12 m by a solid
provides a truly comprehensive and straightforward methodology foundation system made up of reinforced concrete piles and
that can be applied to most common components (cylindrical bracing beams. Finally, the pipes are transversally joined by struc-
bodies, elbows, spheres.), providing critical parameters tural profiles that transmit the weight of the entire system to the
(minimum allowable wall thickness, safe working pressure, safe main pipeline halfway between every two foundations. Fig. 2
moment estimate.) for the corresponding working conditions. shows a scheme of one piece of the model used to make the
When compared to the equivalent methodologies of other well structural analysis of the whole system, showing the geometry of
known procedures some considerations can be made: the structure. It can be observed that there is an expansion loop
introduced in the route of the pipes in order to avoid high stresses
– It is more general than the ASME code (ASME XI, code case N- and/or buckling problems caused by the relatively high tempera-
597-2), focused on nuclear components. The ASME code also tures (þ80  C) of the chemical products and the corresponding
analyses both circumferential and longitudinal failures but it is length increments of the pipes. In total, there are eight expansion
even simpler. Also, and considering the critical safety aspects loops in the pipelines, one every 200 m, as shown in the pipeline
associated to the nuclear industry, it provides more conserva- plan view illustrated in Fig. 3.
tive results through the consideration of the yield stress as the Table 1 shows the diameters and thicknesses of the different
resistant parameter (and not the flow stress as considered in pipes and Table 2 gathers the tensile properties of the main pipeline
FITNET). material [13].
– The API 579 assessment is, at first sight, a more complete
methodology, but it is also noticeably more complex in terms of
presentation, document extension, and the definition and
obtaining of the different inputs. Also, it is mainly focussed on
pipeline systems designed following API standards. If it is
applied to other systems, designed using other procedures/
codes (or whose design bases are not well known), there are
certain input parameters (design criteria, geometry of the
different defects, historical sequence of the measurements,
defect profiles.) whose definition or availability is not always
straightforward. It proposes three different levels of analysis (I,
II and III), level II being the one which is comparable to the
FITNET FFS methodology when external loads are acting
together with the internal pressure. The higher the level is, the
more accurate the results are. This accuracy can be noticeably
higher than that obtained with FITNET FFS when level III is
applied.

2. Geometry and material properties

The pipeline assessed is the main one in a system composed of

five secondary additional smaller pipes whose integrity is not Fig. 1. Example of LTA located in the pipeline.
 S. Cicero et al. / International Journal of Pressure Vessels and Piping 86 (2009) 329–334 331

Table 1
Diameter and thickness in the different pipes.

Pipeline Outer diameter (mm) Nominal thickness (mm)

Pipe 2400 609.6 9.52
Pipe 1200 304.8 8.38
Pipe 800 203.2 8.18
Pipe 600 (x2) 152.4 7.11

the straight stretches and 4.4 mm for the expansion loops, the
corresponding values for every single straight stretch or expansion
loop not being very far from these values. Moreover, the defects
Fig. 2. Scheme of a stretch of the whole structure (model used in the structural
were located in both the intrados and the extrados of the pipe and
analysis). also in the upper and the lower part. Thus, it seems reasonable to
consider one characteristic defect for each of the mentioned parts
of the pipeline and consider that this defect can be perfectly located
3. Structural analysis in the most stressed point of the pipe. If the fitness for service of the
pipeline is demonstrated with these assumptions no more calcu-
The pipeline is subjected to the following loads: lations would be needed, given that a conservative assumption
would produce a safe assessment. The last data required for the
– working pressure (0.8 MPa); analysis were the superficial dimensions of these defects. These
– own weight (7800 kp/m3); were not supplied by the owner of the installation, so another
– weight of the conversion oil (1000 kp/m3); inspection was necessary to determine them. Since the metal loss
– loads transmitted by the secondary pipes (and the fluids in was mainly produced in the external surface of the pipeline,
their interior); a simple visual inspection was enough to estimate that the
– loads transmitted by a footbridge placed on top of the pipeline dimensions w (width, or circumferential dimension) and l (length,
(90 kp/m), as shown in Fig. 1; or longitudinal dimension) could be taken as 300 mm and 400 mm,
– thermal loads caused during the provision of the conversion oil respectively, for elbows, and 500 mm (w) and 350 mm (l) for the
(þ60  C). straight stretches. This assumption is again conservative (since
these are the maximum dimensions measured) but covers both the
Calculations have been performed using the software Tricalc biggest metal loss areas found in the pipeline and the possible
[14] and the results show, as expected, that all the straight stretches interactions between different defects.
are subjected to very similar bending moments and compression Therefore, the problem has been reduced to the assessment of
loads, something that also happens between the expansion loops. two defects: the first one is located in the most stressed point of the
This simplifies the structural analysis, with just two different expansion loops (elbow) and its dimensions are 300 mm for w and
bending and compression diagrams. Table 3 shows the critical 400 mm for l, the remaining thickness being 4.1 mm; the second
combination in both the straight stretches and expansion loops (the one is located in the most stressed point of the straight stretch, w
axes are defined in Fig. 4). In the latter, the critical section corre- being 500 mm and l 350 mm, and the remaining thickness 4.4 mm.
sponds to the elbow. The assessment procedure can be applied once the geometry of
the defects, the stress state and the material properties are known.
4. Stress analysis In this case, the FITNET FFS corrosion module (chapter 9 in the
procedure) has been used. This consists of two types of assess-
Considering the results shown in Table 3, the worst combination ments: (1) stress corrosion cracking and corrosion fatigue, in
of stresses caused by the bending moments, compression forces chapter 9.1, and (2) assessment of local thin areas (LTA), in chapter
and internal pressure is shown in Table 4 for both the straight 9.2. A comprehensive explanation of the first type of assessment
stretches and the expansion loops. These stresses are the inputs can be found in Refs. [15,16], this paper being focussed on the
used in the structural integrity assessment. application of the second one (which is also introduced in Refs.
[15,17]). Fig. 5 [2,18] shows the flowchart for the assessment of LTA.
5. Assessment of LTA following the FITNET FFS The procedure provides the formulation, after which the analysis
corrosion module can be applied, for the following types of components: cylindrical
body, sphere and vessel end, elbow and nozzles. It also provides
The owner used guided ultrasonic longitudinal waves in order to guidance for interaction rules. Finally, for both a cylindrical body
locate the main metal loss areas in the entire pipeline stretch and an elbow, the analysis can be performed to obtain the safe
(1800 m) and then, ultrasonic transducers were used to measure working pressure, the safe working system stress or the minimum
the metal loss in those specific zones. Due to the enormous number allowable remaining wall thickness. Given that for the case ana-
of local thin areas it was decided not to make a one by one lysed here (and for efficiency reasons in the provision process) the
assessment, but to establish a characteristic defect for both the working pressure is fixed at 0.8 MPa, calculations will be oriented
straight stretches and the expansion loops. This decision, which to obtain the minimum allowable remaining wall thickness.
may be considered too conservative, is quite reasonable after
a visual inspection of the pipeline and the results of the thickness
measurements. These revealed a minimum thickness of 4.1 mm for Table 2
Material properties (lower bound values).

Material Yield stress Ultimate tensile Young’s

(syield, MPa) strength (su, MPa) modulus (MPa)
Steel ASTM A333 205 380 200,000
Fig. 3. Pipeline system plan view (1800 m).
332 S. Cicero et al. / International Journal of Pressure Vessels and Piping 86 (2009) 329–334

Table 3 Table 4
Main forces and moments on pipe 2400 . Total stresses (maximum absolute values) due to external loads and internal pres-
sure. The elbow values do not include the aelbow and belbow correction factors.
Fx (kN) My (kN m) Mz (kN m)
Straight stretches 43.1 50.4 75.7 saxial (MPa) shoop (MPa)
Elbows in expansion loops z0 107.4 55.9 Straight stretches 38.5 25.2
Elbows in expansion loops 39.4 25.2

For the elbows (located in the expansion loops), the steps in the " #
assessment (section 9.2.5.5.3 in Ref. [2]) are shown below. An
syield sh; elbow sh; elbow
rsf ELmin ¼ max ; (5)
explanation of the formulation can be found in Ref. [17]. For sA 0:85scyl syield
internal pressure, it assumes that the stress distribution in a pipe
The hoop stress should not be higher than the code allowable
elbow can be approximated by the membrane stress solution to the
stress [2]. Typical code allowable stresses (sA) for pipelines are
thin shell equilibrium equations for an axisymmetric shell of
listed in Table 5. Here, the following value has been used [2]:
revolution [19,20]:  
syield suts
sA ¼ min ; (6)
(a) Determine the allowable remaining wall thickness to prevent 1:5 3
longitudinal failure (hoop stress):
(a.1) Calculate the nominal failure stress of an unflawed pipe (scyl), sA is 126.6 MPa and rsfELmin is 0.18.
following Ref. [21]: (a.5) Calculate the length correction factor (Q):
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 n l2
1 Q ¼ 1 þ 0:8 (7)
scyl ¼ suts (1) Do t
2
where suts is the ultimate tensile strength and n [22]: where l is the length of the defect. Q is 3.66.
(a.6) The minimum allowable remaining wall thickness to prevent
65 longitudinal failure (tmL) is:
n ¼ (2) 
syield
rsf ELmin ð1Q1
syield is the yield stress expressed in MPa. Here, scyl is 305 MPa. tmL ¼ t (8)
(a.2) Calculate the elbow parameter aelbow using:
1rsf Q
ELmin

giving tmL ¼ 1.39 mm.

1RRb
aelbow ¼ R
(3)
12R b
(b) Determine the allowable remaining wall thickness to prevent
circumferential failure (axial stress):
where R and Rb are defined in Fig. 6. For the case analysed, R is (b.1) Calculate the axial bending stress (sz) for moment M (in this
299.9 mm and Rb is 1524 mm and, therefore, aelbow is 0.89. case My caused by thermal loading) of a pipe without
(a.3) Calculate the hoop stress, sh: corrosion:
P ðDo tÞ M
sh; elbow ¼ (4) sz; elbow ¼ (9)
aelbow 2t belbow Z
where P is the working pressure, Do is the outer diameter and t is where
the thickness.
Here, sh is 28.4 MPa. l2=3
b ¼ (10)
(a.4) Calculate the minimum required remaining strength factor 0:9
(rsfLmin) for the nominal hoop stress (sh, elbow):
tRb
l ¼ (11)
R2
 
D4o D4i
z Z ¼
p
(12)
32 Do
Di is the inner diameter. Here l is 0.161, belbow is 0.329, Z is
0.002725 m and, finally, sz, elbow is 119.8 MPa. Here it is impor-
tant to note the influence of the elbow geometry in the stress
state, since belbow provides stress values around three times
bigger than the corresponding ones for a straight pipe.
(b.2) Calculate the axial stress (sa, elbow) from the axial force in the
y bend:

F
x sa; elbow ¼
pDt
(13)

Here, the axial stress is negligible.

(b.3) Calculate the elbow system stress (ssys, elbow):

Fig. 4. Axes definition in the pipeline section.

ssys; elbow ¼ sa; elbow þ sz; elbow (14)
 S. Cicero et al. / International Journal of Pressure Vessels and Piping 86 (2009) 329–334 333

Fig. 5. Flowchart LTA assessment procedure [9.1].

(b.4) Calculate the flow stress (sf): tm ¼ max½tmL ; tmC  (20)

syield þ suts Therefore, for the expansion loops and for the hypotheses
sf ¼ (15) considered here, the minimum thickness required to ensure
2
their integrity is 2.30 mm.
Here, sf is 292.5 MPa.
(b.5) Calculate the minimum required strength factor rsfECmin for
A totally analogous assessment process is provided in FITNET
circumferential failure for the elbow system stress, ssys, elbow:
FFS [2–4] for the evaluation of cylindrical bodies (straight stretches)
  and, therefore, the calculation process is not presented here. The
1 sh; elbow
rsf ECmin ¼ X1 þ X2 ssys; elbow (16) formulae are also analogous, but not including the geometrical
sf 2
factors a and b used for the consideration of elbow geometry. Also,
being the allowable remaining wall thickness is calculated for both the
" # longitudinal and the circumferential failure. For the case analysed
syield sh; elbow here and for the hypotheses considered, tm is 1.90 mm (tmC being
X1 ¼ max ; (17)
sA syield the critical parameter).
!
 sf
X
sSA  1s
syield 2 A
X2 ¼ (18)
sSA s2A
sA is the code allowable hoop stress and sSA is the code allow-
able axial stress, both following Eq. (6) (see Table 5). Here,
rsfECmin is 0.66.
(b.6) The minimum allowable remaining thickness to prevent
circumferential failure (tmC) is given (as shown in Fig. 9.9 in
Ref. [2]) by:
" #
rsf Cmin 1
tmC ¼ t 1 (19)
½0:707cos1 f0:45sin Dwo 0:318Dwo 1:111

For the case analysed, tmC is 2.30 mm.

(c) Finally, the minimum allowable remaining wall thickness (tm)

for combined internal pressure and axial loading is given by: Fig. 6. Elbow dimensions [1].
334 S. Cicero et al. / International Journal of Pressure Vessels and Piping 86 (2009) 329–334

Table 5
Summary of typical allowable design stresses for typical pipelines and piping.

Design code

B31.3 Chemical plant B31.4 Liquid transportation B31.8 Gas transmission and ISO 13623 Petroleum and natural
and refinery piping systems for hydrocarbons distribution piping systems gas industries pipelines
Allowable circumferential minðSMYS UTS
1:5 ; 3 Þ 0.72 SMYS Ranging from 0.40 to 0.80 SMYS depending 0.45–0.83 SMYS depending on
stress on location class location class and product
Allowable total axial stress minðSMYS UTS
1:5 ; 3 Þ 0.54 SMYS 0.75 SMYS for onshore pipeline zSMYS depending on location
for non-incidental loads 0.8 SMYS for offshore riser and pipeline class and product
minðSMYS UTS
1:5 ; 3 Þ
Allowable external stresses 2 0.18 SMYS 0.35–0.55 SMYS depending on location class z0.58–0.77 SMYS depending on
at full design pressure location class and product

6. Discussion As shown here, the FITNET FFS corrosion module provides

a comprehensive and relatively simple methodology for assessing
Once the minimum allowable remaining wall thickness has been components (cylindrical bodies, spheres and vessel ends, elbows
obtained in both the elbows and the straight stretches, the results can and nozzles) with local thin areas. This procedure constitutes an
be compared to the measured values in order to evaluate the struc- alternative to other well known procedures, providing a general
tural integrity of the pipeline and, finally, take the corresponding methodology that covers the four major failure modes (not just the
decisions (i.e, inspection periods, replacements, repairs, etc): LTA assessment) and being applicable to all kinds of industries. As
an example the results obtained using the FITNET FFS procedure
– For elbow locations (placed in the expansion loops), the have been compared to those obtained using the ASME code and
minimum allowable thickness is 2.30 mm, which is noticeably API 579, and the differences detected have been justified.
lower than the minimum measured (4.4 mm). Then, thickness
has been reduced by 5.4 mm in 30 years, which represents References
a metal loss rate of 0.18 mm/year (considering a constant rate)
and, therefore, the minimum allowable thickness would be [1] European fitness-for-service network. EU’s framework 5, proposal no. GTC1-
2001-43049, contract no. G1RT-CT-2001-05071.
achieved in 11.66 years. However, given that w and l also [2] ISBN 978-3-940923-00-4. In: Kocak M, Webster S, Janosch JJ, Ainsworth RA,
increase with time, the actual value would be lower. This fact, Koers R, editors. FITNET fitness-for-service (FFS) procedure, vol. 1. Geesthacht,
together with the uncertainty of the metal loss rate, makes it Germany: GKSS Research Centre; 2008.
[3] ISBN 978-3-940923-01-1. In: Kocak M, Hadley I, Szavai S, Tkach Y, Taylor N,
recommendable to perform the inspection of the elbows editors. FITNET fitness-for-service (FFS) annex, vol. 2. Geesthacht, Germany:
noticeably before the mentioned 11.66 years (i.e, 2–3 years). GKSS Research Centre; 2008.
– For straight stretches, the minimum allowable thickness is [4] Kocak M., Laukkanen A., Gutiérrez-Solana F., Cicero S., Hadley I., editors. FIT-
NET fitness-for-service (FFS) case studies and tutorials, vol. 3. ISBN 978-3-
1.90 mm, which is still somewhat lower than the minimum 940923-02-8, in press.
measured thickness (4.1 mm). The corresponding thickness [5] British Standard BS7910. Guide to methods for assessing the acceptability of
loss rate (if constant) is 0.19 mm/year and the expected time for flaws in metallic structures. London: BSI; 2000.
[6] RCC-MR: Règles de Conception et de Construction des matériels mécaniques
the minimum allowable thickness to be achieved is more than
des ı̂lots nucléaires RNR. AFCEN; 2002.
11.58 years. Again, this value is an upper bound (given that w [7] R5: assessment procedure for the high temperature response of structures.
and l also grow) and the thickness loss rate is uncertain, so it is Procedure R5, Issue 3. Gloucester, UK: British Energy; 2003.
recommended here to perform the next inspection much [8] R6: assessment of the integrity of structures containing defects. British Energy
Generation. Report R/H/R6, revision 4; 2001.
before 11.58 years (i.e, 2–3 years). [9] Bergman M, Brickstad B, Dahlberg L. A procedure for safety assessment of
components with cracks – handbook. SAQ/FoU report, 9/101, AB Svensk
If the LTA analysis is performed using the ASME code (case N- Anläggningsprovning. Swedish Plant Inspection Ltd; 1991.
[10] API 579. Recommended practice for fitness for service. Draft issue 4. American
597-2) and API 579, the results obtained are the following: Petroleum Institute; 1996.
[11] ASME boiler and pressure vessel code, section XI: rules for in-service
– Both the ASME code and API 579 do not provide explicitly inspection of nuclear power plant components. The American Society of
Mechanical Engineers; 1995.
solutions for elbows when external loads are applied simulta- [12] JSME. Codes for nuclear power generation facilities – rules on design and
neously with the internal pressure. Therefore, FITNET FFS construction for nuclear power plants. JSME S NC1; 2001.
provides solutions for a wider range of industrial situations. [13] ASTM A333M-88a/ASME SA333. Seamless and welded steel pipe for low
temperature service.
– For straight stretches, ASME provides a minimum allowable [14] Tricalc 6.4. Cálculo de Estructuras Tridimensionales. Arktec; 2007.
remaining thickness of 2.82 mm, and API 579 provides (consid- [15] Kocak M. FITNET fitness for service procedure: an overview. In: Proceedings of the
ering a reasonable allowable remaining strength factor, RSFa, of 0.8) international conference on fitness-for-service FITNET. Amsterdam; 2006. p. 3–14.
[16] Gutiérrez-Solana F., Cicero S. FITNET FFS procedure: a unified European
2.01 mm. The ASME solution is more conservative, while the API
procedure for structural integrity assessment. Engineering Failure Analysis.
579 solution is very similar to that obtained using the FITNET FFS. Available online, 10.1016/j.engfailanal.2008.02.007.
The main difference between ASME and the other two procedures [17] Ritchie D., Koers R. Assessment of local thinned areas. In: Proceedings of the inter-
is the resistant parameter considered in the calculations: the national conference on fitness-for-service FITNET. Amsterdam; 2006. p. 281–99.
[18] British Standard BS7910:2005. Guide to methods for assessing the accept-
former uses the yield stress and the latter uses the flow stress. ability of flaws in metallic structures; 2005.
[19] Miller AG. The plastic collapse of cracked pipe bends under internal pressure
or in-plane bending. Report no. TPRD/B/0806/R86. UK: Central Electricity
7. Conclusions Generating Board; 1986.
[20] Bickell MB, Ruiz C. Pressure vessel design and analysis. 1st ed. Macmillan and
Company Limited; 1967.
Local thin areas detected in a pipeline placed in a marine
[21] Stewart G., Klever F.J., Ritchie D. An analytical model to predict the burst
environment have been analysed using the FITNET FFS corrosion capacity of pipelines. In: International conference on offshore mechanics and
module. After considering some conservative (but realistic and arctic engineering OMAE. Pipeline technology, vol. 5. American Society of
reasonable) hypotheses, the structural integrity of the system has Mechanical Engineers; 1994. p. 177–88.
[22] Wallin K., Laukkanen A. Theoretical basis for correlation between irradiation
been demonstrated and recommendations about inspection induced change in yield strength and shift in fracture toughness transition
periods have been proposed. temperature. VTT research report BTU072-041309; 2004.