Iai Port Structures

7/27/2019 Iai Port Structures

1/16

Page 1

Seismic Performance and Design of Port Structures

Susumu Iai1, Tetsuo Tobita

2, M. ASCE, and Yukio Tamari

3

1Professor, Disaster Prevention Research Institute, Kyoto University, Gokasho, Uji, Kyoto 611-0011

Japan; [email protected] Professor, ditto; [email protected]

3Senior Research Engineer, Tokyo Electric Power Services, 3-3-3 higashi-Ueno, Taito-ku, Tokyo

110-0015 Japan, [email protected]

ABSTRACT: This paper describes an emerging methodology for seismic evaluation

and design of port structures. The methodology is based on minimum life-cycle cost

principle through probabilistic evaluation of performance. While conventional seismic

design of port structures are based on a particular return period specified for design, themethodology based on life-cycle cost allows for consideration of ground motions with

all (or varying) return periods and uncertainty in geotechnical conditions. Themethodology based on life-cycle cost also allows the probability of failure being

evaluated rather than prescribed by an authority. Since ordinary port structures are for

commercial use, the method based on the life-cycle cost has potential advantages overconventional design. An example is presented. In particular, the simplified design charts

that are based on a series of parametric studies of effective stress analyses have greatbenefit for implementing the proposed methodology in practice. Apart from the

ordinary port structures, higher priority should be assigned for safety as a performance

objective. This issue is especially relevant if port structures are essential parts of

post-earthquake emergency strategies for recovery and restoration of urban areas. The

paper also discusses the importance of this issue and the future direction of study.

INTRODUCTION

Seismic performance of port structures, such as caisson quay walls, sheet pile quay

walls, and pile-supported wharves, are significantly affected by ground displacements

and soil-structure interaction phenomena, and pose complicated engineering problems.

In particular, seismic performance of these structures in the past earthquakes indicatethat deformations in ground and foundation soils and the corresponding structural

deformation and stress states are key design parameters and, unlike the conventional

limit equilibrium-based methods, some residual deformation may be acceptable in

design. Since 1995 Hyogo-ken Nambu, Japan, earthquake (Fig. 1), significant advances

Geotechnical Earthquake Engineering and Soil Dynamics IV GSP 181 2008 ASCE

right ASCE 2008 Geotechnical Earthquake and Engineering and Soil Dynamics IV Congr


2/16


3/16

Page 3

-serviceability during and after an earthquake: minor impact to social and

industrial activities, the port structures may experience acceptable residual

displacement, with function unimpaired and operations maintained or

economically recoverable after temporary disruption:

-safety during and after an earthquake: human casualties and damage to propertyare minimized, critical service facilities, including those vital to civil protection,

are maintained, and the port structures do not collapse.

The performance objectives also reflect the possible consequences of failure.For each performance objective, a reference earthquake motion is specified as follows:

-for serviceability during or after an earthquake: earthquake ground motions that

have a reasonable probability of occurrence during the design working life;

-for safety during or after an earthquake: earthquake ground motions associated

with rare events that may involve very strong ground shaking at the site.

Although these descriptions are very general, they constitute the essential principles of

emerging methodologies for performance-based evaluation and design of port

structures as described in the next section.

EVALUATION OF SERVICEABILITY THROUGH LIFE-CYCLE COST

While safety should be one of primary performance objectives for ordinary buildings,

serviceability and economy become higher priority issues for ordinary port structures.For these port structures, a methodology based on the principle of minimum life-cycle

cost may be ideal (eg. Sawada, 2003). This methodology is emerging and will beeventually adopted as the state-of-practice in the coming decade.

Life-cycle cost is a summation of initial construction cost and expected loss due to

earthquake induced damage. Probability of occurrence of earthquake ground motion (i.e.

earthquake ground motions with all (or varying) return periods) is considered forevaluating the expected loss due to earthquake induced damage. The life-cycle cost also

includes intended maintenance cost and cost for demolishing or decommissioning whenthe working life of the structure ends.

When evaluating serviceability through life-cycle cost, failure of a port structure is

defined by the state that does not satisfy the prescribed limit states typically defined by

an acceptable displacement, deformation, or stress. If a peak ground motion input to the

bottom boundary of soil structure systems is used as a primary index of earthquake

ground motions, probability of failure F ( )F a at peak ground motion a is computed

considering uncertainty in geotechnical and structural conditions. A curve described by

a functionF( )F a is called a fragility curve (Fig. 2(a)). Probability of occurrence of

earthquake ground motions is typically defined by a slope (or differentiation) of a

functionH ( )F a that gives annual probability of exceedance of a peak ground

acceleration a . A curve described by a function H ( )F a is called a seismic hazard curve(Fig. 2(b)).

Given the fragility and seismic hazard curves for a port structure, annual probability

of failure of the port structure 1P is computed as follows:




4/16

Page 4

FF

PGA: a

FH

PGA: a

0.85

Tail end

(a) (b)

FF

PGA: a

Damage degree I

Damage degreeIV

Damage degree IIIDamagedegreeII

FF1FF2

FF3

FF4

(c)

Fig. 2 Schematic figures of a fragility curve (a), a seismic hazard curve (b), and a

group of fragility curves for multiple limit states (c)

H1 F

0

( )( )

dF aP F a da

da

=

(1)

If a design working life is T years, probability of failure of the port structure over the

design working life is given by

( )11 1T

TP P= (2)

If loss due to earthquake induced damage associated with the prescribed limit state is

designated byDc , expected loss over the design working life of a port structures DC is

given by

D DTC P c= (3)

Thus, the life-cycle costLCC is given by adding initial construction cost IC ,

maintenance costMC and demolishing cost ENDC as

LC I D M ENDC C C C C = + + + (4)




5/16

Page 5

This is generalized further by introducing more than one serviceability limit state.

Given the fragility curve defined for the ith

limit state as F ( )iF a (Fig. 2(c)), Eqs.(1)

through (4) are generalized as follows:

H1 F

0

( )( )i i

dF aP F a da

da

=

(5)

( )1i1 1T

TiP P= (6)

D Di Ti iC P c= (7)

LC I D M ENDi

i

C C C C C = + + + (8)

As demonstrated for liquefaction hazard evaluation by Kramer et al. (2006), the

probability evaluated by Eqs.(1) and (2) is a consistent index of hazard and the

conventional approach based on the return period prescribed in design provisions andcodes can be either too conservative or unconservative depending on the site. Expected

loss evaluated by Eq.(3) is an index that reflects the consequence of failure. Life-cyclecost evaluated by Eq.(4) is an index that properly reflects the trade-off between initial

cost and expected loss. The design option that gives the minimum life-cycle cost is the

optimum in terms of overall economy. Thus, the optimum design has a certainprobability of failure given by Eq.(2). This probability is not prescribed by an authority

(such as 10% over 50 years) but rather determined as a result of the minimum life-cyclecost procedure. The probability of failure can be large if a consequence of failure in

meeting the performance criteria, as measured by seismic lossDc , is minor. The

probability can be small, however, if a consequence of failure, as measured byDc , is

significant. Thus, the minimum life-cycle cost procedure reflects the possible

consequences of failure and, thereby, satisfies the principles in performance objectivesin the ISO guidelines described in the previous section.

AN EXAMPLE

A caisson quay wall with a water depth of 15m is considered for a design example.

The quay wall is constructed on loosely deposited sand for replacing alluvial clay layer

beneath the rubble foundation as shown in Fig. 3. The design options considered in this

example include sand compaction piles (SCP) for foundation only (Case A),

cementation of foundation only (Case B), SCP both for foundation and backfill with a

spacing of 1.8m (Case C), the same as Case C but with a spacing of 1.6m (Case D), and

SCP for foundation only and structural modification by expanding the width of caisson

by 3m (Case E).More than one limit states are typically defined for port structures as shown in Table 1.

Limit states as specified by the acceptable normalized horizontal displacement of a

caisson is shown together with the associated repair cost (or direct loss) in Fig. 4. Ideally,all of these limit states should be used for life cycle cost evaluation (Ichii, 2002). For

simplicity, however, only one limit state was assigned for the example in this paper: the




6/16

Page 6

Fig. 3 Cross section of a caisson quay wall considered for a design example

Table 1 Acceptable level of damage in performance-based design*Acceptable levelof damage

Structural Operational

Degree I :Serviceable

Minor or no damage Little or no loss of serviceability

Degree II:Repairable

Controlled damage** Short-term loss of serviceability***

Degree III:Near collapse

Extensive damage innear collapse

Long-term or complete loss ofserviceability

Degree IV:Collapse****

Complete loss ofstructure

Complete loss of serviceability

* Considerations: Protection of human life and property, functions as an emergency base fortransportation, and protection from spilling hazardous materials, if applicable, should beconsidered in defining the damage criteria in addition to those shown in this table.

** With limited inelastic response and/or residual deformation*** Structure out of service for short to moderate time for repairs**** Without significant effects on surroundings

20000

18000

16000

14000

12000

10000

8000

6000

40002000

0

0 0.05 0.1 0.15 0.2 0.25 0.3

Normalized seaward displacement(displacement/wall height) d/H

Degree I:500 thousandyen

DegreeIII:5000 thousand yen

Degree II:1000 thousand yen

Degree IV:15000 thousandyen

Fig. 4 Repair cost (or direct loss) for a gravity quay wall (Ichii, 2002)




7/16

Page 7

acceptable horizontal displacement of caisson wall is assumed to be 10% of wall height

that corresponds to the displacement of 1.5m. Design working life was assigned as 50

years.

Fragility curvesF( )F a were computed through Monte Carlo simulation by

considering the uncertainty in SPT N-values with 1000 trials for each peak ground

acceleration a input at the base rock. Coefficient of variance in SPT N-values was

assigned as 0.1 based on the number specified in the technical standard of Japanese

ports (2006).In the performance-based design, displacement of wall should be evaluated for each

trial SPT N-value. In theory, this is achieved by running an effective stress analysis of

soil-structure systems for each trial SPT N-values. An example of the results ofeffective stress analysis specifically run for the caisson quay wall for a particular

earthquake motion is shown in Fig. 5.

In practice, this is not feasible. In order to overcome this difficulty in practice, a set of

simplified performance evaluation charts were developed through a parametric study

(Iai et al., 1999; Ichii et al., 2002; Higashijima et al., 2006).

The cross sections and primary dimensions used as input parameters are shown in Fig.

6. Typical examples of the results of the parametric study are shown in Fig. 7. These

results were compiled as a comprehensive set of data for the simplified performance

evaluation charts. These charts are incorporated in a spread sheet format. Input data

required are (1) basic parameters defining the cross section of structures, (2)

geotechnical conditions as represented by SPT N-values, and (3) earthquake data, as

represented by wave form, peak ground acceleration, or distance and magnitude from

the seismic source. These charts can be also conveniently used for efficiently assessingthe vulnerability of coastal geotechnical structures that extends a long distance, such as

tens of kilometers, over a variable geotechnical and structural conditions.

The fragility curves computed by using these simplified performance evaluation chartsare shown in Fig. 8(a). Together with a hazard curve at a port in Japan considered for the

example exercise shown in Fig. 8(b), probability of failure (i.e. probability that does not

meet the prescribed limit state) was computed through Eqs.(1) and (2). For example,probability of failure over 50 years was 0.96 and 0.32 for Cases A and B, respectively.

Fig. 5 Residual displacement of a caisson wall through effective stress analysis




8/16

Page 8

(a) Leaning bulkhead

(b) Embankment type

(c) Gravity type

Fig. 6 Cross sections and primary parameters used for the simplified design charts(Higashijima et al., 2006)

0.35

0.1

0.05

0.25

0.15

0.3

0.2

00.1 0.2 0.3 0.4 0.50 0.70.6

Input Excitation Level (g)

(b) D1/H=1.0

0.35

0.1

0.05

0.25

0.15

0.3

0.2

00.1 0.2 0.3 0.4 0.50 0.70.6

Input Excitation Level (g)

N:EquivalentSP TN-value

N=5

N=10N=8

N=25N=20N=15

(a) D1/H=0.0 Fig. 7 Parameter study of normalized horizontal residual displacements of a

gravity caisson wall over varying maximum accelerations (Iai et al., 1999)




9/16

Page 9

(a) Fragility curves

(b) Hazard curve

Fig. 8 Fragility curves and a hazard curve

The loss due to earthquake induced damage is typically defined on the basis of the

acceptable level of structural and operational damage given in Table 1. The structural

damage category in this table is directly related to the amount of work needed to restore

the full functional capacity of the structure and is often referred to as direct loss due to

earthquakes. The operational damage category is related to the amount of work needed

to restore full or partial serviceability. Economic losses associated with the loss of




10/16

Page 10

serviceability are often referred to as indirect losses. In addition to the fundamental

functions of servicing sea transport, the functions of port structures may include

protection of human life and property, functioning as an emergency base for

transportation, and as protection from spilling hazardous materials. If applicable, the

effects on these issues should be considered in defining the acceptable level of damagein addition to those shown in Table 1.

In the example exercise, the direct loss that is needed for restoring the damage

caisson wall was assigned as one million yen/m following the results shown in Fig. 4(Ichii, 2002). Together with the initial construction cost, including the cost for soil

improvement or structural modification, the life-cycle costs considering only the direct

loss were computed as shown in Fig. 9(a). Design option Case B base on cementation of

foundation showed the minimum life-cycle cost whereas design option Case E based on

structural modification showed the most expensive life-cycle cost.

Indirect loss due to earthquake induced damage to port structures needs careful

evaluation in economic loss of industries associated with service of port. In this example,

for simplicity, indirect loss was assumed as five times as the direct loss. The results

shown in Fig. 9(b) indicate that inclusion of indirect loss reflects the difference indesign options and resulting seismic performance much more clearly than when onlydirect loss is considered. Serviceability during and after the earthquake is obviously

closely related to the operational aspect of port structures.

(a) Considering direct loss only (b) Considering both direct and indirectlosses

Fig. 9 Life-cycle cost of a caisson quay wall (per meter)




11/16

Page 11

EVALUATION OF SAFETY

As previously mentioned, evaluation of safety becomes the primary issue for a certain

category of port structures designated as an essential part of emergency bases with

objectives and functions being emergency use. The emergency base is used foremergency cargo and human transport through sea when the inland transports are

seriously damaged (Figs. 10 and 11). The emergency base is also used as a heliport to

allow emergency air transport.For evaluation of safety, scenario type approach of urban infrastructure systems

becomes important. In particular, modal shifts in transportation at emergency may be

one of the important aspects of operation of ports for emergency. As shown in Figs. 12

and 13, air transport is the fastest to respond to emergency transport of humans whereas

sea transport can be an intermediate means of emergency transport when the recovery of

land transport is slow and difficult as in Kobe during 1995 earthquake.

Once the safety is guaranteed, then the design options will be evaluated based on the

life-cycle cost following the procedure described in the previous section. In this way,

the optimal design is achieved for the type of port structures that are essential part ofemergency base.

In seismic hazard analysis for evaluating earthquake ground motions, earthquake

ground motions for safety evaluation may be determined by either probabilistic or

deterministic analysis. This earthquake ground motion can be determined by adeterministic analysis when an active seismic fault is located nearby. A deterministic

analysis evaluates earthquake ground motion by selecting individual earthquakescenarios, including earthquake magnitude, fault location, fault dimension, and source

mechanism. Deterministic analysis does not explicitly consider the probability of

occurrence of earthquakes but it does consider uncertainties involved in the evaluation

of ground motion from a scenario earthquake. The deterministic analysis is widely

adopted in Japan and western coast of U.S.A.(Leyendeckeret al., 2000)

The methods for seismic hazard analysis include empirical, semi-empirical, andtheoretical methods, and a combination of these methods, and are chosen, consistent

with the degree of refinement required for analysis of the geotechnical works, based on

- the importance of a structure, and

- the available information on seismic faults and deep basin structures in the vicinity

of a site.

In the past, either the largest earthquake motion recorded was adopted as design

ground motion, or magnitude of a scenario earthquake and distance to the site were used

to scale the given time history for specifying design ground motions. Those simplified

methodologies were once called the deterministic approach. Probabilistic approach was

introduced in seismic hazard analysis based on the frequency-magnitude relation by

Gutenberg & Richter (1942).

Extensive studies on the earthquake ground motions near a seismic fault haverevealed that, if a focus is directed to the near seismic fault regions, regularities of

earthquake occurrence (i.e. maximum magnitude or characteristic earthquakes,relations between the energy accumulation and release periods of fault activities) can be

recognized in contrast to the random occurrence model as represented by the

Gutenberg-Richter relation. In addition, new methodologies for characterizing the




12/16

Page 12

Fig. 10 Emergency operation of cargo supply through Japanese navy at

Kashiwazaki, Niigata, port at 2007 Niigata-ken chuetsu-oki earthquake, Japan

Fig. 11 Collapse of a highway in Kobe city during 1995 Hyogo-ken Nambu

earthquake, Japan




13/16

Page 13

Fig. 12 Modal shifts for human transport in Kobe after 1995 Hyogo-ken Nambu

earthquake (Takahashi et al., 1997)

Fig. 13 Modal shifts for cargo transport in Kobe after 1995 Hyogo-ken Nambu

earthquake (Takahashi et al., 1997)




14/16

Page 14

strong earthquake motions have been developed by taking into account realistic source

mechanisms including the effects of asperities and source-to-site path through

appropriate numerical methods such as empirical Green's function method. The

discussions on the probabilistic and deterministic approaches made with respect to

ISO23469 reflect these recent developments in seismic hazard analysis and strongearthquake ground motions.

Primary discussions provided for introducing the deterministic approach as an

alternative to probability approach may be summarized as follows (Iai, 2005):(1) The new methodologies for characterizing the strong earthquake motions such as

empirical Green's function method are easily incorporated in the deterministic

approach. Incorporation of these methodologies into the probabilistic approach has

not yet been established for use in practice.

(2) Probabilistic approach explicitly specifies the probability in dealing with the

uncertainty whereas deterministic approach deals with the uncertainty through a

series of engineering and scientific judgments. There seems a misleading impression

that the probabilistic approach does not involve the engineering judgments. In

particular, epistemic uncertainty is considered to be due to a lack of scientificknowledge and is evaluated by processing opinions from a number of experts. A logictree is often used to process the differing experts' opinions in the march to a final

hazard estimate. The weights assigned for differing opinions are also based on the

engineering judgments. The probability explicitly specified in the probabilisticapproach reflects those engineering judgments.

(3) The deterministic approach is applicable and adequate for determining earthquake

ground motions for design especially when uncertainty is small in evaluating the

parameters of the scenario earthquake. In this case, the deterministic approach can be

an adequate method for evaluating a low probability event as an alternative to the

probability approach that typically involves a large uncertainty in the tail end of the

probability function (Fig. 2(b)).

(4) Probability of occurrence of earthquakes at nearby active seismic faults is often lessthan a few percent over 30 years. For example, 1995 Hyogoken-Nambu (Kobe)

earthquake had the probability ranging from 0.4 to 8% over 30 years when the

earthquake occurred in 1995 (Shimazaki, 2001). Depending on the probability

specified for seismic hazard analysis, the effects of seismic hazard due to nearbyseismic faults might be obscured or ignored, resulting in an unconservative seismic

hazard evaluation.

As listed above, there are many issues that should be studied before applying the

probabilistic approach in seismic hazard analysis for evaluating safety requirements.

Depending on the seismic activities around the site and the available research and

investigation results, an appropriate approach, including the deterministic approach

based on the scenario earthquake, should be used when active seismic fault is located

nearby. It would be advisable to avoid the mechanical use of the probabilistic approachfor seismic hazard analysis.




15/16

Page 15

SUMMARY AND CONCLUSIONS

An overview is given on an emerging methodology for seismic evaluation and design

of port structures based on life-cycle cost.For ordinary port structures where primary objectives and functions are for

commercial use, serviceability and economy become high priority issues and the

methodology based on life-cycle cost has potential advantages over the conventionalseismic design. In this methodology, failure is defined as the state where a structure

does not meet the limit state or acceptable damage level. Probability of failure over

design working life is computed based on fragility curve(s) and a seismic hazard curve.

The fragility curve(s) reflect uncertainty in geotechnical and structural conditions. The

seismic hazard curve allow consideration of ground motions with all (or varying) return

periods. The performance is evaluated in terms of expected loss due to earthquake

induced damage that reflects the consequence of failure. Life-cycle cost properly

represents the trade-off between the initial construction cost and the expected loss. The

design option that gives the minimum life-cycle cost is the optimum in terms of overalleconomy.

An example presented in this paper demonstrates that the simplified performance

evaluation charts that are based on a series of parametric studies of effective stress

analyses have great benefit for implementing the methodology based on life-cycle costin practice.

Apart from the ordinary port structures, higher priority should be assigned for safetyif port structures are essential parts of post-earthquake emergency strategies for

recovery and restoration of urban areas as planned by local, regional, or federal

governments. The methodology based on the life-cycle cost still plays an important role

for arriving at the optimum design from the design options that have been already

confirmed to meet the performance objectives of emergency facilities or safety

requirements. There are certain controversial issues that need careful study forspecifying the earthquake ground motions used for safety evaluation.

ACKNOWLEDGMENTS

The authors acknowledge the financial support by NEDO for this study.

REFERENCES

Gutenberg, B. and Richter, C. F. (1942). "Earthquake magnitude, intensity, energy, and

acceleration."Bulletin of the Seismological Society of America, 32: 163-191.

Higashijima, M., Fujita, I., Ichii, K., Iai, S., Sugano, T. and Kitamura, M. (2006).

"Development of a simple seismic performance evaluation technic for coastalstructures." 2006 Ocean Development Symposium.

Iai, S., Ichii, K., Liu, H. and Morita, T. (1998). "Effective stress analyses of portstructures." Soils and Foundations, Special Issue on Geotechnical Aspects of the

January 17 1995 Hyogoken-Nambu Earthquake, 2: 97-114.

Iai, S. (1998). "Seismic analysis and performance of retaining structures." Geotechnical




16/16

Page 16

Earthquake Engineering and Soil Dynamics III Geotechnical Special

Publication No.75, ASCE. 1020-1044

Iai, S., Ichii, K., Sato, Y. and Liu, H. (1999). "Residual displacement of gravity

quaywalls - parameter study through effective stress analysis." 7th U.S.-Japan

Workshop on Earthquake Resistant Design of Lifeline Facilities andCountermeasures against Soil Liquefaction, Seattle, Vol. NCEER-99-0019,

549-563.

Iai, S. (2005). "International standard (ISO) on seismic actions for designinggeotechnical works - An overview." Soil Dynamics and Earthquake

Engineering, 25: 605-615.

Ichii, K., Iai, S., Sato, Y. and Liu, H. (2002). "Seismic performance evaluation charts forgravity type quay walls." Journal of Structural Mechanics and Earthquake

Engineering, JSCE, 19 (1): 21-31.

Ichii, K. (2002). "A seismic risk assessment procedure for gravity type quay walls."

Journal of Structural Mechanics and Earthquake Engineering, JSCE, 19 (2):

131-140.

Kramer, S. L., Mayfield, R. T. and Anderson, D. G. (2006). "Performance-basedliquefaction hazard evaluation: implications for codes and standards." 8th U.S.National Conference on Earthquake Engineering, San Francisco, Paper No.888.

Leyendecker, E. V., Hunt, R. J., Frankel, A. D. and Rukstales, K. (2000). "Development

of maximum considered earthquake ground motion maps."Earthquake Spectra,16 (1): 21-40.

Sawada, S. (2003).A new look at Level 1 earthquake motions for performance-baseddesign of civil engineering structures: Japan Society of Civil Engineers.http://www.jsce.or.jp/committee/eec2/taishin/index.html.

Shimazaki, K. (2001). "Recent progress in long-term earthquake forecasts in Japan."

Journal of Geography, 110 (6): 816-827.

Takahashi, H., Nakamoto, T. and Yoshimura, F. (1997).Analysis of performance of sea

transport mode immediately after 1995 Hyogo-ken Nambu earthquake Port andHarbour Research Institute. Technical Note, No.861


Documents

Iai Port Structures