You are on page 1of 19

130 Journal of Marine Science and Technology, Vol. 11, No. 3, pp.

130-148 (2003)

A SIMPLIFIED APPROACH TO ESTIMATE THE
ULTIMATE LONGITUDINAL STRENGTH OF SHIP
HULL

Hsin-Chuan Kuo* and Jiang-Ren Chang**

Key words: simplified approach, ultimate longitudinal strength, hull girder, “collapse” have been identified as responses to an ex-
stiffened panel. cessive hull girder bending moment. These may take
place, presumably, with but a single application of load
so that the possibility of their occurring must be weighted
ABSTRACT against the extreme or worst likely load value. In
additions, because aging ships may have suffered dam-
In this paper, the calculation of the ultimate longitudinal strength
ages due to hull corrosion and fatigue, their structural
of the ship hull is revisited. A simplified approach dealing with the
longitudinal ultimate strength of ship hull has been developed and an resistance may weaken and further collapse under ap-
effective numerical program is implemented based on the individual plied loads even smaller than designed loads.
components of ship structures. It is, then, possible to find the ultimate The conventional longitudinal strength calcula-
strength of the whole ship structure by evaluating those individual tion for a ship mainly concerns the elastic response of
components. Comparisons of numerical results from the proposed
hull girder due to an assumed wave load and then, the
approach and published literatures are conducted for further validation.
Since relative errors of calculations of the current approach to the elastic section modulus of hull girder, which is taken as
literature with respect to the experimental data are within 3%, it is the primary surveying index of the longitudinal bending
shown that the proposed approach can be adopted for the estimation strength, can be achieved by using the linear bending
of the ultimate bending strength of ship hull in sagging or hogging theory. However, this criterion may fail when two ships
condition with and without at fully plastic state. In addition, a
having identical section modulus with the same materi-
comparison between the numerical result and the existing expression
for a very large crude carrier also shows that the proposed approach als may have different bending strengths due to differ-
is not only as accurate as those published literatures but also more ing resistance to buckling of the compressed parts of
available. Therefore, the proposed approach with the numerical girder sections.
program is quite suitable to be adopted in the preliminary design of With the full scale structural tests, Vasta [44]
ship structure due to its accuracy and simplicity.
found that a close correlation between the ultimate
bending strength of the hull and the buckling strength of
INTRODUCTION
the compressed parts of the hull plating exists and
proposed that the limiting bending moment of longitu-
In the past three decades, a great deal of activity
dinally framed hull should be the product of the section
has taken place for predicting, in a realistic manner, the
modulus and the ultimate stress of plating panels. This
sea loads to which the ship hull girder is subjected. The
formula is simple; however, it takes no account of
main initiative behind these activities is the possible
buckling effects from compressed parts. With a com-
deficiencies in the standard calculations of the longitu-
parison to the conventional elastic bending theory,
dinal bending moments, particularly under certain
Caldwell [5], therefore, developed the “limit design”
circumstances. The ultimate strength of ship structures
ideas and introduced the concept of a structural instabil-
has been considered very important in the societies of
ity strength reduction factor for strength calculations.
classification and academic research since the loads
However, only calculation with a statical condition of
acting on the ship hull are uncertain due to rough seas or
loading may not be conformity with the real dynamic
unusual loading and unloading of cargoes during
loading of the girder. In additions, although the buck-
operations. Several potential forms of catastrophic
ling effect has been included in calculation, it may not
be permissible to replace the distribution of longitudi-
Paper Submitted 06/17/03, Accepted 08/12/03. Author for Correspondence:
nal compression by an equivalent average ultimate lon-
Jiang-Ren Chang. E-mail: cjr@sena.ntou.edu.tw
Associate Professor, Department of System Engineering and Naval gitudinal compression since the buckling phenomenon
Architecture, National Taiwan Ocean University, 2, Peining Road, Keelung, depends on the stability of structures and has little or no
Taiwan 202. connection with a “mean stress”. Nevertheless, when

Gordo et al. either tensile or estimating average stress-strain characteristics up to compressive. Faulker et al. Cladwell [5] could be the first one to incor. However. which does not neglect any reserve of found [5]. mentioned works with introducing new stress-strain [13] proposed a reasonable method to calculate the relationships. a modification to the formulae proximating method with a rationally based structure will enable the ultimate moment of resistance to be design concept. was a hybrid method of the finite element scheme with ened panel in comparisons with experimental results by the load-deformation curves. By incorporating buckling behav. however. Although some applications [4. The accuracy of this method. Smith [42] also pro. Later. The ultimate longitudinal strength opinion. which proposed to explore the ultimate moment of the stiff. Kuo & J. mainly based on the calculation. in this method. finite element programs. The computing the derivation of moment-curvature relationship for a program was ever validated by calculating the ultimate complete hull. corrosion and combined bending for the ship girder were focused on the collapse strength of illustrating its capability. Adamahak [1] developed approximate assessment of the ultimate longitudinal an approximate method in a computer program and strength of the hull girder. Special attention was also driven to the collapse and posed the incremental curvature procedures to calculate post-collapse behavior of these panels. column behavior. was also reported [41] and empirical curves might lead to inaccuracy. those algorithms. by comparing its predictions with those of Ref. moment of ships. in our charge of oil in 1980. assessment. buckling and post-buckling behavior of compressed tion and the plate element strength reached by way of parts of the same actual ship. [10] to develop ways of claimed that limit states of these panels. H. and lack of accurate modeling of the hull element collapse to the next was suggested instead of girder results in unreliable results even in the case of an applying regular small increments of curvature. The method was also checked obtained curves of moment-curvature for the calcula. Since then. Based on a flexural-tor. Hughes [25] proposed a simpler ap- searches in this field. Besides. By adopting most positive features of the above- Following with Caldwell’s research. since the ultimate strength of the girder section is mainly ior models into an overall analysis of the hull girder as governed by the behavior of panels under compression. which put emphasis on modeling of the finite element formulations. Therefore. They ened panels inspired Dow et al. the bending curvature increases. 41] and hull girder by the linear superposition of individual especially with respect to plate strength and beam- contributions of elements. mance of each individual element for the hull girder and in order to estimate the ultimate longitudinal bending as the numerical capability of a computer grows fast. It should be usually differ one another due to their different noted that. element was proposed to calculate the strength of the Following with previous studies [4. used for calculating the els makes the evaluation of the ultimate strength of a ultimate longitudinal strength of the ship hull girder. some calculating schemes. [17] also presented an sional buckling formulation. redundant structural system impossible. ship or a box girder. 11-13. panel compressive strength until its maximum resis- porate the buckling effects into the ship strength tance has been reached. The proposed method reveals an individual panel structure and the ability to predict sound results as compared to those obtained by using moment-curvature relationships. ignoring such a contribu- the collapse strength and their post-collapse behaviors. Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 131 the buckling characteristics are found with more re. discrete steps from one assumptions. experimental investigation [32]. 8] of this longitudinal bending moment of a very large crude approach have demonstrated its general feasibility with carrier (VLCC) which “broke its back” during dis- comparisons to model experiments. and reliable results and it first led the optimal concept to ship estimates [18. 30]. 43] have been emerged for analyzing structural designs.R. to calculate the perfor- introducing a dynamic relaxation method [29]. were modeled in an appropriate manner compressive strain. tion of the post-collapse resistance of compressed pan- However. 18.C. the assessments on the longitudinal strength of residual stresses. need much time consump. 19. . showed that both the calculation and provision of longi- An engineering approach [4] considering a simpli. at this the VLCC and then further used to study effects of stage. subsequent progress has been based mainly on the panel strength analysis. was dem- made in the fields of buckling as well as final collapse onstrated by a comparison with several experimental studies of structures [6. 19. a computing methodology for the ultimate value of the proved formulations became available for critical stress longitudinal bending moment at any cross section of a corresponding to various modes of compression failures. tudinal strength are in excess of normal requirements fied analysis model for each individual beam-column for the ship in both hogging and sagging conditions. one stiffener with its associate effective plating. The cross section was modeled as Progressive research on post-buckling strength of stiff. Rahman and Chowdhury [40] developed structural instability strength reduction factor and im. [41] for tion of ultimate strength of panels. two simplified expressions were further Since Smith [42] first proposed an approach. 22.

in this paper. 29. 23. 4. for many implemented soon. MT. 31. The second is that. and twisting moment. when the 34.132 Journal of Marine Science and Technology. 10. causes for failure of ship hull compo- nents can be summarized as: 1. [16] since they were proven impractical for will reduce and how to monitor the longitudinal struc- the use of calculations at a design phase. residual strength after collision and grounding [39] in a ment-curve relationship of the ship hull [17] and the short time is necessary especially when the emergency prediction of the method need to be validated by various rescue is requested and following operations will be experimental results [15]. 1. as shown in Fig. 16. force. However. QV. 46]. 40. a simplified and accu- types of ships. No. 31] to deal with the collapse of hull dinal strength of the ship hull. 37] and it can sumption that the strength of ship hull can be repre- accurately accounts for the load shortening contribution sented by the linear superposition of individual contri- of each plate and stiffener assembly. 41]. and overall buckling of the grillage. the excessive compres- sive stress leads to buckling failure mode since the buckling stress of components is lower than yielding stress of structural material and /or the initial conditions. including of a real VLCC example are also conducted to Basically. such as the initial imperfections. the excessive tensile stress results in yielding failure mode owing to elasto-plastic behavior of the used material. tensile stress and shear stress. therefore. 8. when three important issues were emphasized. one can consider and the current approach is not only simple but also accurate categorize all the loads into the vertical bending moment. the buckling failure mode of web frame only concerns . Although most structural designs for merchant ships THEORETICAL BACKGROUNDS mainly concern their bending strengths owing to ratios of their geometrical dimensions as a slender structure. M H . strength for bulk carriers and the torsional strength for 38. such as the shear methods of components [1. 35. The last is that butions of elements [4. Therefore. tow basic assumptions should be noted in the followings: First. horizontal bending moment. 17. 1. torsional buckling or tripping of panel stiffeners. enough when applying it in the preliminary phase of M V . 3. 24. Vol. containers with large openings. flexural buckling of panel stiffeners. to evaluate or assess the contribution of each structural element to the mo. In their paper. have existed. vertical shear ship structural design. Fig. As a result. the combined effect of the vertical and rate approach to estimate the longitudinal strength of horizontal bending moments needs to be considered ship hull is convenient as compared with the finite carefully since it deals with the collapse of ship hulls element method (FEM). 11. 36. In general. the structural strength Gordo et al. 2(a) to 2(c). 2. as the loading types change and oblique wave jected to the internal and external loads can be grouped into the compressive stress. the shear stress leads to shear buckling failure mode. based on the as- under the combined load effects [14. 3 (2003) alternative methods in the field to predict the ultimate arises. Several comparisons section under combined load effects. Besides. residual stress and other more sensitive boundary conditions. For simplification of analysis. 47]. The results show that and wave loads operates at sea. the latter were ever criticized by age of an available ship increases. tural strength becomes vitally important. the stresses arise due to ship hull sub- however. when a ship subjected to the loadings validate the proposed approach. the first is that the ship accident happens [45. Buckling failure modes It is known that there exist four typical modes of buckling failures for ship structural members as shown in Figs. Different loads applied to the hull section. of those are buckling of a plate. Moreover. a simplified considering the case of biaxial compressive strength of approach with a self-developed numerical program is plates [20-21] leads to the proposal of the interaction proposed to deal with problems of the ultimate longitu- equations [14. 41] and the finite element formulations [2. the other kinds of strength appear and play a strength can be grouped into simplified methods or crucial role in the preliminary stage.

as shown in usually exist large transverse frames along the ship hull Fig.5 φu = σ u = (c 1 + c 2λ s2 + c 3β 2 + c 4λ s2β 2 + c 5λ s4) . E Fig. 4. . β. By using the least square method. Faulker et al.R. Kuo & J. such as by Faulker [12]. σ y denotes the aver- age yielding stress of the plate-stiffener combination. The self-developed numerical program of this paper is implemented by using the empirical formulae and we will give a brief introduction for the reader’s convenience as follows. the width of a plate was considered effec- tive and the section of the plate with stiffeners and the simple supported boundary conditions as well as the given loadings were described as shown in Fig. σ x. and Caelsen [6] and (b) empirical formulae by Lin [29]. Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 133 the plate buckling and stiffener’s buckling since there subjected to axial compressive stress. y (1) σy σy Fig. λ s. Paik et al. σ u denotes the t E E ultimate stress of stiffened panel.C. H. respectively. Among them. Failure modes of beam-column components. 5. 3 and their buckling modes can be represented by such that no tripping effect arises. The numerical calculation results are obtained by two parameters. with stiffeners attached to the transverse frame are Approaches used to analyze those buckling failures Based on assumptions mentioned above. 1. and Lee [27]. the avail- able approaches have been proposed and can be further grouped into (a) the semi-empirical formulae. λ s = πar . second. Lin’s formula (without effective breadth) Lin [29] first proposed the dynamic relaxation method to derive the reduction factor of the ultimate strength of a plate with stiffeners and adopted the beam- column element to analyze the strength of ship structure. A transverse frame attached with stiffened panel. and of a plate. in which β = b . 4. Several buckling failure modes for different structural members. the buckling reduction factor containing the two param- eters can be formulated by σ – 0. 3. Paik and Pedersen [36]. those plates the beam-column components as shown in Fig. the slender ratios of a stiffener. Fig. Paik and Mansour [35]. [13]. 2. [37].

Approach used to analyze yielding failure When the structural members are subjected to yielding loads.067. Based on the assumption.765. 11. Paik et al. 6(b).09 displacement method to deal with the nonlinear equilib. (5) As + Ap in which A p and A s represent the sectional areas of a plate and a stiffener.156λ s4 0 ≤ λ s < 1. Approach used to analyze shearing failure As for the buckling failure due to shear forces existing in a panel member of rectangular section be- tween two stiffeners as shown in Fig. An analytical model for the sectional area of a stiffened panel when the initial imperfection existing at the center of with boundary conditions. 4. Vol. respectively. Paik’s formula in which Based on Lin’s work and with more numerical f = 1 + 0. if there exists no crack in member.209λ s2 + 0. is adopted. In his work. φu = σ u = 1 (1 + 0. σ – 0. b the width of a plate. respectively. (4) rium equation of the elasto-plastic large deformation of a plate with stiffeners. formula of the buckling reduction factor as f = λ s2 λ s ≥ 1.5 φu = σ u = (c 1 + c 2λ s2 + c 3β 2 + c 4λ s2β 2 + c 5λ s4) . an assump- tion should be noted that the sectional area of a plate with stiffener remain unchanged until the buckling of the panel subjected to shearing effects yields.16 0. c 2 = 0. 5. Fig. Therefore.936. λs. The yielded as five constants in equation (1) are c1 = 0. c 4 = 0. c2 = 0.131.59 data. in general.48 0. σ y = 0. Lee’s formula Davison [9] proposed an empirical formula to deal Following with Lin’s research.59 σ – 0. Davison’s formula 3.23 + β – β 2 + β 3 . respectively. c 4 = 0. the buckling reduction factor can be thickness of a plate. then all of them ing deformation from the Euler beam theory and the can be used to calculate the buckling failure of a plate without stiffeners as equation (4). c 3 = 0. t the tion.134 Journal of Marine Science and Technology. displacement function is adopted to deal with the bend. As for the estimation of that of a plate. the average yield- ing stress used in the paper for the plate with stiffeners can be yielded as σ yp ⋅ A p + σ ys ⋅ A s σ ya = .170. [37] proposed the similar expressed . No. a three-order It should be noted here that once the parameter. the following formula in which c1 = 0.995.188. the rectangular panel and the simple supported bound- .046.15β 2) . respectively. the elasto-plastic behavior can be illustrated as Fig. in equations (1) to (3) is set to be zero. c 5 = 1.176. 6(a). a the length of stiffeners’ span linear function is considered for the in-plane deforma- between transverse girders. and r the radius of gyration.960. σyp and σys denote the yielding stresses of a plate and a stiffener. c 5 = −0. All the formulae listed above are used to calculate y (2) the buckling failures of a plate with stiffeners. (3) y f 2. Lee [27] adopted with the buckling failure of a plate as the beam-column model and further applied the matrix σu 1. 3 (2003) the Young’s modulus. therefore.5 c 3 = 0.

Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 135 ary conditions given at the edges of the panel are both in which the elastic shearing stress. and then to calcu- late the bending moment by giving the curvature of section gradually. 6 0. (8) τ K 22 – 4K 1K 3 • K – φu τ = τ u = 0. H. based on the simpli- fied procedure. that is. Lin [29] further proposed a regression Fig. respectively. for b' ≥ 1 kτ = (9) critical shearing stress.5 in which α = a' . in which σ ya (b). section. K2 = 0. the preprocessing with modeling and numerical calcula- tions needs much time and expenditure such that a fast assessment before an immediate rescue operation or that at a preliminary design stage of ship hull becomes unavailable. τ E. basically. a n d λ = [1 + 81 + 4τ E y b 81(1 + α 2)4 625 2 0.5 81 ( 1 + α ) + 81 (1 + α ) ] . the FEM is available. for τ E ≤ 0. these researches constructed the relationship of bending moment and curvature.R. parameters t and b' are illustrated in Fig. (a) The elasto-plastic behavior of the structural member.0 ( a ) . the shear reduction factor can be yielded as as [37] τ E = kτ σ E. Since Billingsley [4] proposed a simpli- fied calculating procedure to fix such a problem. for a < 1 2 elastic shearing buckling effect and is obtained by modi. to calculate the ultimate bending strength of ship hull. Above all. however. then. Based on beam theory. Kuo & J.34 + 4. 6.9 ⋅ 1 . can be expressed considered.0126τy. It should be noted that the 5. the initial imperfection and residual stress are consid- ered at specified plate with stiffeners of each member to construct the stress-strain diagram. More specifically. many researchers have begun to develop the simplified ap- proaches and carried out several experimental models to validate their accuracy and reliability. k τ is defined as τy = and σ ya is the average yielding stress as de 3 2 scribed in equation (5). the constants. for τ E ≤ 0. (7) 32α 2λ τy τ α4 τ y(1 – ). Finally the ultimate bending moment can be determined by way of the diagram of bending moment and curvature. σ E = π 2E ( t )2 where E and υ 12(1 – υ) b' in which τ u and τ cr denote the ultimate and critical are the Young’s modulus and the Poisson’s ratio. to deal with such a problem. b' b' fying the plastic coefficient.084τy. in which a reduction factor exists to represent the shear forces existing in a panel member of rectangular the popular and frequent used members of ship structures. K 1 = respectively. τ cr . τE. k is selected the smaller 2 2 2 τ 25 1 + 9α 2 25 9 + α 2 among calculated values of the two equations. shearing stresses. (b) formula. to calculate the ultimate bending strength of a box . Up date.34 + 4.C.1067τy + τcr.0 (b' a a ) . (6) cr 2K 1 In equation (8). or τ π2 . and K3 = 1. a kind of nonlinear elasto-plastic behavior of structural members usually with initial imperfection and residual stress.5 kτ = (10) τ cr = y . Methods to calculate ultimate longitudinal strength The problem of calculating the ultimate longitudi- nal strength of ship hull is. is yielded based on the 5.

ultimate horizontal bending moment. 7. calculations of bending moments only concern nor- mal stresses such as buckling and yielding effects. 2. while those of shear and free torque effects consider shear stress and shear flow. of which several useful formulae describing the ulti- mate longitudinal bending moment strength under sag- ging and hogging conditions and the ultimate horizontal bending moment were derived. 11. 7 to 9. the members are independent of each other when calculating the various ultimate strengths of members. Shear stress distributions for a container. Paik and Mansour [35] further proposed an improved simplified method. No. the section of a calculating member is unchanged when subjected to loadings. and post-buckling effect is neglected. Fig. the compressive and tensile load- ing conditions adopt the ultimate stress and yielding stress. shearing stress with related bending moments. Following with Lin’s idea. respectively. those formulae for calculating ultimate vertical bending moments under sagging and hogging conditions. (b) shear stress distributions for a tanker. respectively. the ship is floating at upright condition.136 Journal of Marine Science and Technology. the bending strength Fig. 8. Therefore. . In their methods. shearing stress and torsional stress due to free torque for different types of ship’s sections as shown in Figs. 4. 1. ulti- mate shear strength and free torque assuming that no longitudinal constraint exists in the closed channel of ship section are revisited and can be yielded as follows. Vol. sec- tions of structural members subjected to loadings are assumed unchanged. The basic assumptions for used simplified calculating procedures are: 1. (a) Normal stress distributions for a tanker. The present paper follows Paik et al’s simplified procedures and further considers different buckling and yielding stresses. Ultimate vertical and horizontal bending strengths Based on the beam theory. 3 (2003) girder. 3.

M Vus formulae for calculating the ultimate vertical bending moments of sagging and hogging conditions and ulti. For lengthy of the paper. 11. stress distributions can be illustrated in Fig. for calculating the vertical mate horizontal bending moment are described in the bending moment in sagging is following. H. The coordinate system of a hull section for calculating the stress distributions for a bulk carrier. 11. 9. Fig. . Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 137 only concerns the normal stresses. The normal stress. only used condition. Kuo & J.C. Detailed derivations of those formulae can Fig. 10. therefore. σ x. ultimate vertical and horizontal bending moments. 10 and the related coordinate describing the hull section can be (1) Ultimate vertical bending moment at sagging shown in Fig. Normal stress distributions for ultimate vertical and horizontal bending moment conditions. all the be found in Lee’s work [26] for the interested readers. (a) Shear stress distributions for a bulk carrier.R. (b) torsional Fig.

for y = y 1 (outer bottom) (15) loads due to the normal stress applying on the section is equal to zero and thus. . therefore. for y = y m = 0 (at deck) approaching to the neutral axis as shown in Fig. 10(b). and σ uD represents the ultimate buckling expressed as stress of deck. the ultimate vertical bending moment in hogging can be yielded as [35] . No. for y = y m condition. (16) uSL + σ ySU n (σ uSU + σ ySL ) Σ Azj j =1 (13b) and and H = C 1D + C 21D 2 + 2C 2D . g=σ . 3. (m – 1) H (2) Ultimate vertical bending moment at hogging – σ uD .138 Journal of Marine Science and Technology. and σ ulB as well as σ uB represent the ultimate σ ySL ⋅ H buckling stresses of the inner and outer bottoms. – σ uSL . (m – 1) H sentative member between the deck and neutral axis. the ultimate vertical bending moment in m–1 sagging can be yielded as [35] m–1 Σ A y i (D– y i ) i =3 M Vus = – A ymσ uD (D – g) + 1 (σ uSU + σ ySL ) H Σ A y i (gy i – i =2 y 2i ) C2 = Σ n Azj . therefore. 10(a). It should be noted that the two param- eters have a linear relationship when they are gradually σ yD . and σ uSL represents the compressive stress of a representative member between the inner in which bottom and neutral axis. can in which σ yD represents the mean yielding stress of the be yielded as deck. for 0 ≤ y ≤ H H – (H – 3g) σ ySL ] (14) σx = – σ uSU . i = 2. A yi and A zj represent the coordinates of (17b) horizontal and vertical members and their correspond- ing area. σySU represents the yield- ing stress of a representative member between the deck H = C 1D + C 21D 2 + 2C 2D . – 1 [(σ uSU + σ ySL ) y i – H σ ySL ] . for H ≤ y ≤ D . (13a) and neutral axis. The position of neutral axis... and σ uSU represents the compressive stress of a repre. σySL represents the yielding stress of a repre. in which σ yB represents the mean yielding stress of the σ x. for H ≤ y ≤ D σx = . H for 0 ≤ y ≤ H sentative member between the bottom and neutral axis. respectively. the normal stress. Vol. thus. it means the summation of normal – σ uB . – 1 [(σ uSL + σ ySU ) y i – H σ ySL ] . and D the height of section. σ xdA = 0 . g. for y = y 2 (inner bottom) Besides. – 1 [(σ uSL + σ ySU ) y – H σ ySU ] . – σ ulB . the position of neutral axis. 3 (2003) n σ yB . 4.. for calculating the vertical bending moment can be outer bottom.. Since a linear relationship exists when they are gradually approaching to the neutral axis as and shown in Fig. and Therefore. ... (17c) j =1 m–1 n + σ ySL Σ A y i (y i – g) i =2 – 1 2D Σ Azj j =1 (D – H) (D Therefore. (17a) m–1 Σ (A y i y i ) i =2 in which C2 = n . z j. M Vuh (11) Again at the hogging condition. (12) uSU + σ ySL respectively. for y = y i . (13c) Σ n m–1 j =1 Azj (σ uB A y + σ ulB A y + σ uSL 1 2 Σ A z – σ yDA y m – σ ySU i Σ= 3 A y i ) j =1 j C1 = n (σ uSL + σ yUL ) Σ Az j =1 j in which y i. be yielded again as (σ uD A ym + σ uSU Σ A z j – σ yB A y 1 – σ ySL Σ A y i) σ ySU ⋅ H j =1 i =2 C1 = g=σ . g. n m–1 can. for y = y i . i = 3. for y = y 1 = 0 (outer bottom) + H – 2g) σ uSU – A y1σ yB ⋅ g – H 6D Σ Azj j =1 [(2H – 3g) σ uSU – 1 [(σ uSU + σ ySL ) y – H σ ySL ] . 11.

the at outer starboard side (at z = z n). oil tanker and bulk carrier. 7(b). (n – 1) 2. 8. it is found that the maximum shear in which σuS and σulS represent the compressive stresses stress almost exists in the side shell or longitudinal of representative members locating at port sides inner bulkhead as shown in Figs. and 9(a). σuH and σyH represent ultimate vertical shearing strength can be expressed as the stresses of those between port to neutral axis and starboard to neutral axis. The in which τ u denotes the ultimate shear stress and A V position of neutral axis can be easily yielded as represents the summed sectional areas of side shell and longitudinal bulkhead. Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 139 M Vuh = – A y 1σ uB (D – g) + A y 2σ ulB (D – y 2 – g) and n –1 m–1 Σ A z j (B – z j ) – 1 (σ uSL + σ ySU ) Σ A y i [g(D – y i ) – (D – y i )2] C2 = j =3 m . therefore. for z = z j . for z = z 2 (port side inner) – σ uS . as shown in Fig. j = 3. (21c) H i =3 Σ Ayi i =1 m–1 – σ ySU Σ A y i (D – y i – g) i =3 – A y mσ yD ⋅ g in which B represents the breadth of the section. be yielded n as [35] – 1 ( Σ A z j ) (D – H) (D + H – 2g) σ uSL 2D j = 1 M Hu = – A z1σ uS (B – g) + A z2σ ulS (B – z 2 – g) n – H ( Σ A z j )[(2H – 3g) σ uSL – (H – 3g) σ ySU ] n –1 6D j = 1 – 1 (σ uH + σ yH) H Σ A z j {g(B – z i ) – (B – z i )2} j =3 (18) n –1 (3) Ultimate horizontal bending moment. m the stress distribution for calculating the ultimate hori. Ultimate shear strength H – σ ulS . .• (23) them have a linear variation relationship when approach- ing to the neutral axis as shown in Fig. σ yS these parts of hull sections are mainly considered to represents the yielding stress of those members locating calculate the ultimate shearing strength and thus. Ultimate free torque strength C1 = m (σ uH + σ yH ) Σ Ayi i =1 It is known that for a ship with large hatch openings. for z = z 1 (port side outer) Based on the numerical calculations of shear dis- (19) tributions of hull sections for the container. . for H ≤ z ≤ B σx = . the torsion rigidity is mainly attributed to and contrib- (21b) uted by its inner closed channels. (21a) bottom shell and the closed deck. • (24) in which in which A H represents the summed sectional areas of H = C 1B + C 21B 2 + 2C 2B . and outer (at z = z 1 and z = z 2 ). 10(c). – σ yh Σ A z j (B – z i – g) j =3 – A z n σ sD ⋅ g Based on the same procedures illustrated above. respectively. m n –1 σ uS A z 1 + σ ulS A z 2 + σ uH Σ A y – σ yH j Σ= 3 A z j – σ yS A z n i =1 i 3. H.R.. (20) uH yH Q uH = τu AH. for z = z n = 0 (starboard side outer) – H 6B Σ Ayi i =1 [(2H – 3g) σ uH – (H – 3g) σ yH] – 1 [(σ uH + σ yH ) z – H σ yH ] . 9(b). Kuo & J. Similarly. 4. for 0 ≤ z ≤ H H (22) – σ uH . – 1 [(σ uH + σ yH ) z j – H σ yH ] .. the ultimate horizon- σ yH ⋅ H tal shearing strength can be represented as g = σ +σ . thus. since both of Q uV = τ u AV. respectively. The ultimate horizontal bending moment can..C. – 1 2B Σ Ayi i =1 (B – H) (B + H – 2g) σ uH zontal bending moment can be expressed as m σ yS .

5 0 2 . As for those representative shear force and free torque. their line of channel walls for the i-th channel. 3 (2003) therefore.5 0 0 .5. stress for fully plastic horizontal bending moment is when λ s ≥ 1. the fully plastic capacity ratio parameters. comparisons of those formulae are conducted 4. under fully plastic state. φ u.0 0 λs Fig. the initial imperfection of the stiffener.140 Journal of Marine Science and Technology. Therefore. 5) of calcu- setting the area of upper part of the cross sectional to the lated stiffened panel. At the range of λ s ≤ 1. the representative yielding contours of the three formulae are similar. According to the suggested dimensions and initial The fully plastic capacity of hull section is usually condition of stiffened panel from British Standard [7] considered for different hull materials.5 0 1 .15% of neutral axis of the hull section is first determined by the longitudinal length (a. that the residual stress. however.5 0 1 . 4 . as shown in Fig.0 0 3 . the contour plots of Paik’s formula determined from the maximum yielding stress at the disperse unreasonably as shown in Fig. .0 0 0 . It is found that to estimate yielding state and the representative yielding stress for the reduction factor. is 0. Vol. 13 to 15. each other are not clear and seldom found in the literature. they are determined from the 5 . δp. is 20% of the yielding stress.0 0 1 . δ s .0 0 0 . in this and Det Norske Veritas (DNV) rules [28] and assuming paper. the vertical and horizontal bending moment.0 0 maximum shear stress of the related shear section and the maximum shear flow existing in the closed channels.5 0 2 . 14 and are quite side shell component. 12. λ s and β for three formulae as de- state is reached when the stresses for upper and lower scribed in equations (1) to (3) can be obtained and areas of the neutral axis are both approaching to the illustrated in Figs.0 0 0 .5 0 4 . shear force and free torque.0 0 β 2 . and qu denotes adaptable ranges of applications and comparisons of the ultimate shear flow of the representative plate. that is. 29. no tripping effect exists. Lin’s and Lee’s formulae are fully plastic vertical bending moment is. as shown in Fig.5% of the put into implementation. obtained more similar especially when the slender ratio of a by choosing the maximum yielding stress value at the stiffener. 11. Fully plastic capacity of sectional area and the results are discussed in the following. 5) of stiffened panel. n As mentioned in the above section.5 0 3 .0 0 3 . λ s . ≤ 1. An effective neutral axis of a sectional area for the fully plastic state. the present paper only concerns the free COMPARISONS AND VALIDATION torque due to the sectional closed channels without longitudinal constraints and it can be represented as a Comparisons of various formulae for buckling failure sum of strengths of all the inner closed channels. is 0.5 0 3 .0 0 1 . thus.0 0 2 . Similarly. three formulae M uT = 2 Σ Ωi q u . as shown in Fig. the deck component.0. then. the effective neutral axis equal to that of lower part to the contour lines showing the relationship of two slender effective neutral axis. (25) [27.5 0 4 . σr.59. 12. therefore. the initial imperfection of the plate. Fig. No. are σy. the fully plastic bending moment. Then. 13. For the calculation of vertical breadth (b. 31] for calculating the buckling reduction fac- i =1 tor of stiffened panels and one [9] for that of a plate have in which Ω i represents the area enclosed by the center been proposed as a simplified approach.0 0 4 . Contour lines of λs and β for Lin’s formula.5 0 5 . However.

0 0 0 .0 0 2 .5 0 1 . Comparisons of the reduction factors of four empirical formulae.5 0 3 .0 0 β 2 . λs 3.C. β 3 . formulae of Lin’s. 2.0 0 0 .5.5 0 1 .0 0 ing with side shell or bulkhead. 17.5 0 3 . A z − y coordinate system as shown in Fig.5 0 3 . 4 .5 0 merical program are necessary and summarized as follows: 2 .5 0 4 . H. Fig. formulae of Lee As for comparisons of the reduction factors of a and Paik are quite similar while at β ≥ 1. 16. it is suitable at the range of β ≥ 1.0 0 3 . However. stiffened plate members and plate mem- 1 . Kuo & J.0 0 1.0 0 λs Fig. Numbering end joints of a member should correspond to the z − y coordinate system.0 0 3 . It should be noted that those areas of the deck.0 0 3 .0 0 0 . Fig.0 0 2 . Contour lines of λs and β for Paik’s formula.0 0 some preprocessing procedures for the developed nu- 2 . those of Lin’s plate.5 0 stiffened plate are considered as corner member.0 0 0 . further compare with that of Davison’s formula as de.5. when β ≤ 0.0 0 4 .5.0 0 1 .0 0 in the hull section. 15.5 0 bers as shown in Fig.5 0 4 . setting every λ s in equations (1) to (3) to be zero and Paik’s are approaching together and Lee’s formula can be used to compute the plate’s reduction factors and shows the largest among the reduction factors.5 0 0 . . formula becomes inadequate to adopt especially when Paik’s.5 0 2 . 5 .0 and comparison results show that its reduction factor is the 5 . Numbering all members and classifying them with suitable type labels.5 0 Validation of the proposed simplified approach 4 .5.0 0 1 .5. As for Davison’s their respective formulae are illustrated in Fig. All calculating results from greater than 1.3 ≤ β ≤ 4. Contour lines of λs and β for Lee’s formula.R. approaching together.0. 14.0 0 Preprocessing and main frame of the numerical program 3 . Members in hull section can be categorized into cor- ner members. 11 is set up 0 . inner and outer bottom shells join- 1 .0 0 1 .0 0 4 .5 0 5 . 16 and it formula. When β ≤ 1.5 0 1 .5 0 2 .0 ≤ β ≤ 1.5 0 To make use of the proposed simplified approach. the reduction factor from Lin’s formula is scribed in equation (4). and two end points of 0 .0 0 4 . 4. It means Paik’s is shown that when 1.0 0 0 . Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 141 different from those of the other twos.5 0 4 .5 0 2 . it seems unreasonable.0 0 smallest among all when 1.5 0 5 . and Lee’s show the same tendency and nearly λ s ≥ 2.

Besides. Hull section members are categorized as corner. MSD represents Model of Ship Double-bottom. as comparing both of them with the experimental data. The flow chart of the main program. all input data can be shown graphically to check positions of nodal points of the model. Geometry of every type of members is illustrated in Fig. and results with comparisons for buckling strength are shown in Table 2. Since the Excel has the graphic function. 20. approach proposed by Beghin et al. 11. Relative errors of calculations of the present approach to the Beghin [3] with respect to the experimental data are all within 3%. MSB repre- sents Model of Ship Bulk Carrier. . With sagging and/or hogging conditions. The flow chart of the main program is shown in Fig. respectively. Those experimental models with di- mensions are illustrated in Fig. The calculating and output sections are implemented by the FORTRAN language.142 Journal of Marine Science and Technology. 18. 19. Vol. The geometry of the calculated member. Comparisons of numerical results with experiments To validate the proposed simplified approach. there are eight kinds of testing models for calculations. 19. Fig. All the needed input data are processed by the Excel software and stored in an xls format. However. In this figure. 17. stiffened panel and plate components. results from MSB model with sagging and hogging cases both show a dominant Fig. Characteristics of material used in the experimental models are listed in Table 1. a simplified Fig. and MSC represents Model of Container. model label MST represents Model of Ship Tanker. No. 18. ex- periment results conducted by Nishihara [33] are used for comparison. 3 (2003) 5. It is shown that the present approach for calculating buckling strength at fully plastic state is as accurate as Beghin et al. [3] from results compared with each other. [3] is also included for further comparison.

In Table 4. have been made in the present Besides. from Lin’s empirical formula (at about 0. Material characteristics used in experimental models t (unit: mm) σy (Kg/mm2) E (Kg/mm2) v thickness of palte yielding stress Young’s modulus Poisson’s ratio 3. therefore.368φu2 . Kuo & J. respectively. to the present methods.R.60 32. Paik and Lee into consideration.461φu2 . those methods including the present one need to be modified when dealing with the buckling strength calculation of a bulk carrier. It should be noted that the present ap- proach takes different reduction factors from Lin.172 + 1. More specifically. M Vuh = 0.3 2.1 2. Such a phenomenon is due to considerations of initial Fig. due to the un.11 × 104 0.35 26. respectively.6% error of uniform scantlings of the hull section. Results of ultimate bending moment calculations as well as com- parisons with the literature are listed in Tables 3 and 4.11 × 104 0. for sagging condition MP . The empirical formula is also used here for comparison and described in the following: M Vus = – 0.289 Note: The material used in experimental models is mild steel. 20.C. then Nishihara’s the MSB model is a typical kind of hull section arrange.9 2. geometry of hull section members is very because some important assumptions. Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 143 Table 1.71%). it is remembered that parameters λ s and β methods.277 4. H. in our opinion. Four experimental models with their corresponding geometry. the residual stress.12 × 104 0. Such an error is. respectively.459φu – 0. imperfection and residual stress in the methods and it makes the original structural members weakened espe- cially for cases 6 to 8. (1. In table 3. and exist in the corner of hull section of a bulk carrier. Lin [29] also proposed an empirical formula to calculate the ultimate vertical bending moment for hull section of box-type. originally describing the slender ratios of a stiffener and a plate. such like the sensitive to the results especially when the wing tanks initial imperfections of the plate and stiffener. now are quite hard to describe the behavior like a wing tank structures. it does not mean the present methods fail strength. for hogging condition MP (26) in which φ u and M P represent the reduction factor and the vertical bending moment. . there can have three alternatives for selection in the present method to calculate the ultimate vertical bending moments. Therefore.113%) and Beghin’s (5. ment of a bulk carrier and for calculation of the buckling however. it is shown that results from the present methods are accurate for cases 1 to 5 and most of them are little smaller than the experimental data except for case 5. average only) are the most accurate.548φu – 0. although those results error.003 + 1.05 29.281 5.

96 1.669 0.264 1.021 0.20 103.071 1.608 1.0 104.58 91.20 103.5 110.005 1. 41.5 61.95 1.28 5 49. The geometry of also used to compare with those from previous studies stiffeners at hull section of this ship is shown in Fig.0)h 2 58.94 51.76 70.0 89.89 58. data [33] (ton-m) (ton-m) (ton-m) Experiment Experiment 1 MST-1sag/hog 94.995 7 0.000 3 1.144 Journal of Marine Science and Technology.915 8 MSC-hogging 88.6 58.63 59.33 61.8s/(95.22 59.951 1.94 45. Calculation results of ultimate vertical bending moments Model Experimental Nishihara [33] Béghin [3] Present with Present wth Present with Lin’s empirical No. Vol.215 4 0.2 49.792 0.5 74.98 89.794 0.2 64.014 1.5 93.43 76.44 84.88 69.5 86.91 73.005 0.0 84.907 0.82 68.30 8 88.2 69.764 1.8 87.65 57.8 59.619 4 MSD-hogging 85. those of others are quite closed to . 11.980 0.8 89.918 0.257 6 1.990 1.185 1.60 88. No.948 0.059 0.1 86.952 0.5 97.797 0. h: hogging condition.60 88.2 45.8 58.923 0.03 1.9 0.00 109.97 1. The output moment in hogging.31 76.10 77.5 92.853 0.1 64. fully plastic vertical and horizontal bending practical ship. 3 (2003) Table 2. 35] and they are listed in Table 6. “Energy Concentration”.164 1.285 7 MSC-sagging 113.9 95.671 0.0 89.951 5 1.9 1.5 96.080 1.78 49.2 57.30 97. The principal dimensions of the ultimate free torque of hull section. Comparisons of various ultimate vertical bending moment calculations Model Nishihara [33] Béghin [3] Present-Lin Present-Paik Present-Lee Lin’s formula No / Experiment / Experiment / Experiment / Experiment / Experiment / Experiment 1 0.672 8 0.1 52. It is shown Two hundreds and eighty five elements with 291 nodes that except for the result of ultimate vertical bending are used in modeling for calculations.8s/(65.181 Table 3.56 90.1 58. These results are VLCC are illustrated in Table 5.893 0.971 1.003 3 MSD-sagging 60.50 4 85. ultimate vertical and horizontal shear stresses.671 0.789 0.90 58.3 76.017 1.5 80.127 0.93 45.800 0.012 1.30 97.986 0.983 1.945 1.03 1. a real case study of a VLCC.793 6 MSB-hogging 68.70 6 68.163 2 MST-2sag/hog 58. fully plastic vertical and horizontal shear stresses and parison and validation.890 0.962 0.906 0.8 103.146 5 MSB-sagging 49.045 1.001 1.138 1.0)h 3 60.960 1.005 2 1.793 0.958 Validation of the proposed approach with a VLCC results from the self-developed program include ulti- mate vertical bending moments with hogging and To further validate the proposed approach for a sagging.092 0.96 1. 21.84 89. Table 4.5 97.29 Note: s: sagging condition.5 104. in [41] is used here for com. [17. namely moments.060 0.16 7 113. Results and comparisons of ultimate strength at vertical fully plastic state Model Model Label Experimental Béghin [3] Present Béghin [3]/ Present/ No.29 56. data [33] (ton-m) (ton-m) (ton-m) Lin (ton-m) Paik (ton-m) Lee (ton-m) formula (ton-m) 1 94.48 81.0 77.184 1.

2% except for the case of ultimate vertical bending moment in hogging (12. our thanks also go to Professor W. such as estima- tions of residual strength of ships after grounding or collision and the corrosion strength assessment of aging ships. Comparisons of ultimate strength calculation results obtained from the proposed approach with experimental data and those from pub- lished literatures have shown that the current approach is acceptable for estimations of ultimate bending mo- ments in sagging or hogging with and without at fully plastic state. 21.894 tons Deadweight 216. CONCLUSIONS In this paper. Kuo & J. .221 H. Besides. further studies. Principal dimensions of a VLCC VLCC “Energy Concentration” Length. thus. financial support under Contract Number: USDDC. summer extreme 19.269 tons Machinery Steam turbine those of published literatures. (a) The geometry of hull section of a very large crude carrier [41].597 m Gross tonnage 98. In addition. H. Wang for some valuable discussion on this work. 41. It should be note that the calculating errors of the present approach to Paik’s approach [35] are within 5.C. 35] since it can be used to calculate other related longitudinal strengths. (b) the geometry of stiffeners at hull section of a very large ACKNOWLEDGEMENTS crude carrier [41]. A comparison of buckling strength calculations with ex- perimental data and published literatures has shown the adaptability and limitation of those available formulae used in ship structural design. between perpendiculars 313.75 m Length.28%). overall 326.R. molded 48.19 m Depth.0 m Breadth. Fig. In addition.2 m Draft. a simplified approach combined with a numerical program has been successfully developed to predict the longitudinal ultimate strength of ships. molded 25. be used as a fast-evaluated tool to estimate the longitudinal ultimate strength of ships especially at the preliminary stage of structural designs due to its simplicity. the current approach is more available as com- pared with the literatures [17. can be carried out based on the current approach. The proposed approach with the available program can. A further comparison of the current ap- proach with published literatures for a practical very large crude carrier has not only validated the proposed simplified approach but also shown its accuracy. Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 145 Table 5. The authors would like to express their thanks to the United Ship Design and Develop Center for their -T452.

9. D. Hugill. 485. “The Ultimate Longi. Compression.C. pp. Computer Code for Evaluation of the Ultimate Longitu. and Vetter. Imperical College. 5. K.” In Proceedings. 1-17 (1975). Caldwell.. 1695-1715 (1980). Beghin. and Walker.. “Design of Plate Panels Under Biaxial 2. 51-63 (1980). C. 507 (1993). Coronado. “Collapse Compression.K. 11. 221-246 (1973).D....” Design.8402E+12 1. - capacity of hull Fully plastic vertical 5. No.B. Engin. Interna. D. - shear stress of the hull Ultimate free torque 4.55E+12 1. Chen. Bending Moments. England (1989). 7. PRADS95.26E+07 .10833E+11 . 107.” In Proceedings. 5th Star Symposium. Davidson. D.. ASCE. Inter- REFERENCES national Symposium on Practical Design of Ships and Mobile Units. Arlington.” In Proceedings. R. 3.. - shear stress of the hull Fully plastic horizontal 6. Vol.64E+12 hogging condition. “Ultimate Longitudinal Strength.C. P.S. 3.31E+12 - bending moment Ultimate horizontal 2.” RINA the Analysis of Stiffened Plating in Bending and Trans. pp. Vol. and Smith. Struct.843 (1995). Vol.90E+12 1. “An Approximate Method for Esti. .” J.. “Hull Girder Response to Extreme Stiffened Plate Panels. 11.445E+12 .C.W. Extreme Loads Response Symposium.. 2.. M. pp. Jastrzebski.32E+07 . SNAME. pp. “A Review of Effective Plating for Use in 5. T.. .745-2.S. P.” In Proceedings. 411-430 (1965). [35] Gordo et al.P.12E+07 - shear stress of the hull Ultimate horizontal 6. Lembit.19E+12 - Bending moment Ultimate vertical shear 4..832-2. Dow. J.. Ship Res. Vol.. Analysis of Ship Hulls. Mobile Units.9807E+12 1. Elsharkawi. and Petersen.. Fully plastic vertical 2. 133-147 (1981). 33-40 (1980). Compression and Normal Loads. 6. Vol. G.. C.” J. ened Plates in Compression. .S. 2.” Computers Struct.. 6.” Ph. Bendiksen. [17] approach Caldwell [41] Ultimate vertical bending moment 1. 12. . and Taczala. - bending moment Fully plastic horizontal 3.” In dinal Strength of Hull Girder.W.” Marine Struct. J. Vol.. Faulker.3561E+12 . 37-61 (1984). 2. 8.92E+12 of the hull in sagging Ultimate vertical bending moment of the hull in 1..3604E+12 . C. tional Symposium on Practical Design of Ships and SNAME.. D.... Symposium.916E+12 1. Y. pp. “Evaluation of Ultimate Strength Ship Hull Strength.. pp.. University of London. pp... 149-169 (1983). . pp. P. J. 91. pp.43E+07 . Carlsen. “Buckling of Multibay 4. and Smith. Adamchak.A. and mating the Collapse of a Ship’s Hull in Preliminary Maciej. SNAME Trans. A. PRADS92. Proceedings. Calculation results of the present approach and comparisons with previous studies Present Ruthrtford and Paik et al.146 Journal of Marine Science and Technology. - capacity of the hull Note: unit of stress: kN.765 (1992). M.T. 106. Faulker..14E+07 . D. tudinal Strength of a Ship Hull.. “Synthesis of Wedged Grillages to Withstand pp.. Y. Snyder. E.B. 56B.C. Div. Christopher. Adamahak. pp.639E+12 1. J.K.J. Billingsley. Bai. “Renault 10. R. . SNAME. M. Chalmers. Shear and Lateral Pressure. unit of moment or torque: kN-m. M.” Struct. 3 (2003) Table 6.F.. Clarke. 1. “Ultimate Strength of Ship Structures. Vol..D. Dissertation.C. Vol. 3. “A Parametric Study of Collapse of Stiff. 13.. Ship Struct. 19. pp.

“Influence of Local Compression Failure on Glasgow.. 154. J.. Soc. pp. C. and Mansour A. 24. Marine Compression.836 (1992).. pp. 9. J. Sekihama. Keelung. A. 471-494 (1996). Smith.. Vol. J. “A Simple Formulation for for Stiffened Plates Under Predominantly Uniaxial Predicting the Ultimate Strength of Ships.T. Res. 449-470 (1996). Proceedings. Vol.E. 40. Department of Naval Architecture and Method to Evaluate the Hull Girder Collapse Strength. 98. Hughes.L. 844-856 (1995). J.” RINA Trans. 13-17 (1983). New York (1983).. Y. Lin.. C. Moments.. 37... pp.. Lin.M. and Che. Vol.” J. Gordo. Vol. 31. PRADS92. and Guedes Soares. Vol. PRADS 95. Guedes Soares.. Thayamballi.” In 28. Guedes Soares. Vol.. M.. J. Taiwan (2000). 30. 230-240 (1997). pp. pp. pp. and Chowdhury. “A Simplified Method 22... 10.. 29. Vol. and Thayamball. 133. Hansen. sium on Practical Design in Shipbuilding.. Kuo & J. pp.. Nishihara. Ship Res. Scotland Marine Struct. ment of Ship Hull Girder. “Reliability Analysis of Continuous Structural dinal Strength of Ships: A Case Study.H. 36. pp. T. Ship Res. 21. “Ship Longitudinal Strength.K. Vol.F. 154-165 25. 35. Bull. under Combing Longitudinal Bending and Shearing 259 (1996). 465-497 Sci. pp...” J. Vol. Ship Res. “Charts for the Buckling and Post-buck- tion for the Collapse of Tankers and Containerships ling analyses of Stiffened Plates under Combined under Combined Bending Moments. pp. Rahman. Mansour... J. “Ultimate proximate Assessment of the Ultimate Longitudinal Strength of Ships under Combined Vertical and Hori- Strength of the Hull Girder. Rutherford.. International Symposium International Symposium on Practical Design of Ships on Practical Design in Shipbuilding. PRADS 95. pp. Vol. ing Corrosion.. Seoul. Strength. “Ap.. C.” J. Guedes Soares. H. 2.” Ph. and Yang.B. 40. 30. Horizontal Bending and Shearing Forces.. Paik. Mansour. 43. Vol.” John 39. S.. “Ultimate 23. J. 38-54 (1998). pp. (1997).” Ocean Engineering. “Approximate Dissertation. 2. Guedes Soares. pp.K. C.E. 411-439 (1990). J. Paik.. Ship Res. 139-158 Compressive Loads. S. J.. “Analysis of Ultimate Strength of Stiff- Under Unaxial Compression.S.. Systems.” SNAME Trans. T. and Yang. and Gordo. 19. Shipbuilding Prog.S. C. “Collapse of Ship ment of Double Skinned Hull Structures against Ulti- Hulls Under Combined Vertical and Horizontal Bending mate Bending and Fatigue Strength.K. Shipbuilding Prog. pp. (1996). sidual Strength Assessment of Ships After Collision and 26.T. Computer-Aided. Gordo. 244-257 (1996).” Marine Struct. Struct. “Interaction Equa.. and Gordo. 15. International Symposium on Practical De- . pp. Scotland (1989). pp.K.” Marine Strength of Ship Hulls under Combined Vertical Bending. O. University of 42. J.K... “Estimation of Ultimate Longitudinal Strength Grounding. C.” SNAME Trans. pp.M.H. “On the Reliability Assess. 104.K.. International Sympo- 69 (1996). J. S. J. Vol. Steel Res.E. J. 12.. A.” J. Optimization Approach.. and Pedersen. 199-215 (1991). “Ultimate Longitu- 27. A. P. and Caldwell.” Int.. 217-228 (1994). 2. “An Imperfections. A... “Hull Collapse of An Aging Bulk Carrier Lateral Pressure.K. H. Kim. Box Girders. S. J.D.. pp.M. Guedes Soares. Strength of Rectangular Plates Under Biaxial Load and 34. and Guedes Soares. pp. 60. Lee. Lee. 287.834-2. pp. 16. Yang. Hori. Paik. Thayamballi. Y. pp.. “Estimation of Department of System Engineering and Naval Ultimate Longitudinal Bending Moment of Ships and Architecture. Construct.K. Technol. Trans. “Structural Trans. 1.C. 367-375 (1983). A. Loading. Department of Naval Vol. C. Based. “Design Equation for Ship Plate 33.” SNAME 24.” J. “Re- Wiley and Sons. C.H. “Behavior and for Predicting Ultimate Compressive Strength of Ship Design of Stiffened Plates Under Predominantly Panel. “Compressive Japan..” J. (1998).” Ph. Gordo.. D. 42.S.. Thayamballi.. ened Rectangular Plate-on the Ultimate Bending Mo- Vol..” In Proceedings. 18.E.. Steel Res. 40... Design by Analysis Approach Applied to a Product Oil 38.. pp.” In Proceedings.” Thin-Walled Struct.” RINA Trans. Vol.. 9. (1976). 31-59 (1996).D.. Lee.. D.M..K. Vol. SNAME. J. “Design Equation for the Compres.. Experimental Investigation of Ship Hull Ultimate 310 (1988). 136. Dissertation.K. SNAJ. National Taiwan Ocean University.M. sive Strength of Unstiffened Plate Elements with Initial 32. 231. C. Vol. J. S. P. Guedes Soares. and Mobile Units. Paik. S. Gordo.R. 9. and Faulker. Paik. Force. Construct. “Ship Structural Design: A Rationally. 441-471 (1990). Ultimate Longitudinal Strength of a Ship Hull. Architecture and Ocean Engineering.. 99-114 (1992). 17-25 41.” J. M. pp. Vol. “Design Methods 35. Vol. pp.H. J. 41.” J. (1985). J.” In Proceedings. Vol.K. 52-62 (1995).M. Naval Architects 20. 98. Chang: A Simplified Approach to Estimate the Ultimate Longitudinal Strength of Ship Hull 147 14. “Ship Hull Ultimate Strength Reliability Consider- Vol.” Tech. Seoul. 808-819 (1995).C. C. A.” Int.. and Soreide. Paik. Vol..” Marine Technol.. Mansour A. No. and Yang.M. M..M... and Paik. and Rashed.. and Risk Assessment for Ship Hull. “Strength of Midship Section.K. zontal Moments. J..H.” Master Thesis. Carrier with Unidirectional Girder System. University of Glasgow. and Guedes Soares.. 2-22. A. 17.

H... “Analysis of the Accident Nakhodka. and Tests. Tokyo. Vol. Vol. Vasta. T. Vol. I..148 Journal of Marine Science and Technology. 11. 437-446 (1992).I. Estimation of Structural Strength. P. 165 (1958).” J. H. pp. Part 1.. No. Japan. Technol. Technol.” J. Soc. pp. Yao.” Com. Sueoka. Vol. H. Marine Sci. puters Struct. “Lessons Learned from Full Scale Structural 47. T.. 729-740 (1980). 181-193 (1998). Vol. Estimation of Wave Marine Sci. Y.. PRADS77. 12. Watanabe.” SNAME Trans. “Analysis of the Accident of the MV 45. pp. 3.. 172.. “Numerical Design Prediction of the Ca.. “Progressive Collapse Analy- pacity of Plates in Biaxial In-plane Compression. pp. 3 (2003) sign in Shipbuilding. 44. Loads. Vol. 43. S. SANJ. 171- . 73-79 (1977).. A. of the MV Nakhodka. Takemoto. 3. Kumano.. and Ohtsubo. and Nikolov.” J. 66. pp. J. pp. Valsgard. Sumi. sis of a Ship’s Hull Under Longitudinal Bending. 46. Naval Archit. SANJ. 180 (1998). Yao.. Part 2..