Ship Design I

Manuel Ventura
mventura@mar.ist.utl.pt

MSc in Naval Architecture and Marine Engineering

Summary

Ship Dimensioning
Owners Requirements
Traditional approach
Generic Ship Dimensioning Process
Most common implementation methods:

Systematic parametric variation

Optimization methods

Some Optimization Software Tools

Excel Solver
Matlab fmincon() function

Annex A.

Ships Statistical Data Gathering and Processing

Annex B.

Physical Limitations to ship dimensions

Annex C.

Economical Measures of Merit

M.Ventura

Ship Dimensioning

Ship Dimensioning

Preliminary Design Process

The determination
of the main
dimensions and
characteristics of
the ship is the first
step of the
preliminary design
stage.
Yang & al (2006)
M.Ventura

Ship Dimensioning

Owners Requirements
The starting point is a set of Owners requirements defining
mainly the ship type, cargo capacity and speed
Example of requirements:
Type of ship:
Mission:
Deadweight:
Max. Draught:
Cargo capacity:
Service speed:
Autonomy:
Cargo Equipment:
Other:

Container-carrier, with cell guides

Service Line Setubal - Antwerp
9,500 dwt
8.0 m
750 TEU, including 20 reefers
17 knots
20,000 miles
2 cranes of 40 t x 26.5 m
Accommodations for 15 people

M.Ventura

Ship Dimensioning

Ship Dimensioning Traditional Process

DW (input)
Assumed a (DW/ Displacement) ratio empirically
Displacement = DW / (DW/ Displacement)
Lpp = f (Displacement, Vs )
Cb = f ( Fn, Displacement, Vs )
B, T, D are functions of:
Space requirements (cargo and ballast volumes, max.
dimensions)
Intact stability
Free Board
Reserve of flotation
M.Ventura

Ship Dimensioning

Selection of the Form Coefficients

Selection of the Cb

In diagrams similar to the one in the figure, as a function of the Froude

Number

M.Ventura

Ship Dimensioning

Selection of the Main Dimensions

M.Ventura

Ship Dimensioning

Ship Initial Dimensions

Watson and Gilfillan (1976) presented the following
procedure to obtain the main dimensions of a ship with the
required displacement

( ) ( )

3
L 2 B

B
T
L=

(
)
1
.
025
1
s
C

L
B=
L
B
B
T=
B
T
D =T D
T

( )

( )
( )

M.Ventura

The ratios (L/B), (B/T)

and Cb are obtained
from statistical data of
similar ships

(D/T) is initially assumed

as 1.20

(1+s) is a coefficient
related to the hull
appendages

Ship Dimensioning

Ship Dimensioning

10

Generic Ship
Dimensioning
Process

Modern approach

M.Ventura

Ship Model
Type of Ship
CDW, TEU, Lane Length
Vs
Autonomy
Etc.

Lpp, B, D, T, Cb
Etc.

Design
Variables

Mission
Requirements

Ship Model

Possible
Solution

M.Ventura

Type of Propulsion System

Specific Fuel Oil Consumption
Etc.

Technical
Design
Parameters

Displacement
Cm, Cwl, Kb, Lcb, BMT, BML, Sw
Lightship Weight, Kg, Lcg
GMt
DW, CDW
Cargo Capacity
Ballast Capacity
Propulsion Power
Length of Engine Room
Length of Cargo Area
Etc.

Ship Dimensioning

11

Size Measures for Specific Ship Types

Weight based design (Ex.: Tankers, bulk-carriers,..)

Cargo capacity depends mainly of the displacement

Homogeneous cargoes
CDW is the measure of cargo capacity
Depth = f(Vcargo)

Volume based design (Ex.: Container carriers)

Unitized or packed cargo
Number of TEU is the measure of cargo capacity

Area based design (Ex.: Ro/Ro ships)

Lane length for vehicle stowage is the most common measure of
cargo capacity
M.Ventura

Ship Dimensioning

12

Constraints

M.Ventura

Ship Dimensioning

13

Economical Assessment

M.Ventura

Ship Dimensioning

14

Economical Measures of Merit

Initial ship cost
The initial ship cost is not by itself a good indicator, some
design options only become economically advantageous on the
long run

Other criteria can be used to take into consideration the

running costs of the ship along its entire operational life
The most common are:
Required Freight Rate (RFR)
Present Value (PV)
Internal Rate of Return (IRR)

To evaluate these criteria the knowledge of the typical ship

voyage is required
M.Ventura

Ship Dimensioning

15

Typical Voyage
The specification of the typical ship voyage allows a more
comprehensive analysis of the economic aspects
It may include:
The number of ports visited during the round trip
The distance between ports
The cargo-handling capabilities available and the corresponding
handling rates and costs
Port fees and taxes

M.Ventura

Ship Dimensioning

16

Example of a Typical Voyage

Specification
Itinerrio
1
2
3
Carga
1-2
2-3
Ritmos de carga/descarga
1
2
Termos de carga/descarga
Custos porturios do navio
1
2
Custos porturios da carga
1
2
Frete
M.Ventura

Setbal
Anturpia
Sines
600 teus x 14 t
400 teus x 16 t + 200 teus vazios
60 teus/hora shinc
70 teus/hora shinc
Li-Lo
10,000 + 0.5xGT
30,000 + 0.5xGT
100/teu cheio, 50/teu vazio
120/teu cheio, 70/teu vazio
RFR (frete mnimo requerido)
Ship Dimensioning

17

Some Common Freight Conditions

fio

(free in and out)

fiost

(free in and out stowed and trimmed)

li-lo

(liner in and liner out)

shinc

(Sundays and holidays included)

sshex

(Saturdays, Sundays and holidays excluded)

M.Ventura

Ship Dimensioning

18

Other Generic Requirements

Ship Registry (conventional flag/convenience flag)

MAR

Duration of the Investment (ship economic lifetime) 20 years

Capital Interest Rate (bank loans)

10%

Working days /year (Off hire days/year)

355 d (10 d)

M.Ventura

Ship Dimensioning

19

Generation Engine for Design Variables

M.Ventura

Ship Dimensioning

20

10

Determination of the Design Variables

Parametric Studies
The independent variables are obtained by variation between
the lower and upper limits assumed
Require more computing time when the number of design
variables is high
No guarantees that the solution found is the optimal

Optimization Methods
The independent variables are obtained from an optimization
algorithm
Possible to find a better and faster solution
Only provides information about the optimal point found (single
objective methods)
M.Ventura

Ship Dimensioning

21

Parametric Studies

11

Parametric Study Methodology

System of 5 equations
11 variables
3 variables fixed based on
the Owner requirements
(DW, CCAP, V)

= L B T Cb
= LWT + DW

LWT = f ( L, B, D, T , Cb, PMCR )

PMCR = f ( L, B, T , Cb,V )
D = f ( L, B, Cb, PMCR , CCAP )

Experience shows that these three relations,

which are relatively stable for each ship type,
are suited for a good initial estimate
Introducing these additional three relations, the solution of
the displacement equation it is transformed in the solution of
system of eight non-linear equations.
M.Ventura

Ship Dimensioning

23

Ship Dimensioning

24

Functional
Diagram of the
Dimensioning by
Systematic
Variation
The system of nonlinear equations is
solved by an iterative
process

M.Ventura

12

Main Dimensions
The main dimension can be obtained from the ratios and
coefficients used as independent variables
For example:
The ratio k can be obtained
from statistics:

1
= DW
k

DW
k =

T=

L B
Cb
B T

B
B = T
T
L
L = B
B
M.Ventura

Ship Dimensioning

25

Optimization Methods

13

Types of Optimization Problems

Single-Objective

Simplified process in which one only objective, considered the

most important, is selected

Multi-Objective

Closer to the reality

Several objectives can be in conflict between them

Hybrid

A multi-objective problem is transformed into a single

objective
One of the objectives is selected as the most important and
the other are converted into a set of constraints that are
varied parametrically

M.Ventura

Ship Dimensioning

27

Types of Methods and Algorithms

Linear Methods:

Linear Programming (LP)

Newton

Non-Linear Methods:

Gauss-Newton
Levenberg-Marquardt
Sequential Quadratic Programming (SQP)
Artificial Neural Networks (ANN)
Genetic Algorithms (GA)
Simulated Annealing (SA)
Particle Swarm Optimization (PSO)

M.Ventura

Ship Dimensioning

28

14

Non-Linear Methods (1)

Linear Successive Approximations

The process starts from an initial feasible point

The functions are expanded in Taylor series around the initial
point, considering only the linear terms
The constrains and the objective functions are linearized in a
similar way and the problem is solved as linear.

Random Search

The values of the design variables are generated randomly

between the lower and upper limits.
The values that do not comply to the constrains are eliminated
and are not used in the next functions.
The process stops when all the variables comply to the criteria
defined.

Direct Search

The process starts from an initial point and generates a

sequence of point that converge to an optimal point where the
function has a minimum.

M.Ventura

Ship Dimensioning

29

Non-Linear Methods (2)

Advantages/Disadvantages
Linear Successive Approximations Fast process but where
the non-linear behavior of the relations is lost due to the
linearization of the initial stage.
Random Search Slow process where the optimum point can be
missed due to the contraction process. It can be applied to
multi-modal functions.
Direct search Based on local search and global movements.

M.Ventura

Ship Dimensioning

30

15

Types of Optimization Methods

Global
Is able to search through the entire design space to find
the optimal solution
Local
Can converge to a local solution, missing possible solutions in
other regions of the design space

M.Ventura

Ship Dimensioning

31

Some Optimization Software Tools:

Excel Solver

16

EXCEL Solver (1)

Available algorithms:
LP - Linear Programming (assumed only if selected in <Options>)
Non-Linear Programming (assumed by default)
GRG2 - Generalized Reduced Gradient (Lasdom et al, 1998)
The Solver approximates the Jacobian matrix (partial
derivatives) using finite differences and re-evaluates it at
the beginning of each iteration

Limits

1 objective (Single Objective algorithm)

200 variables
100 implicit restrictions
400 simple restrictions (upper/lower limits)

Usage:
<Tools>/<Solver>
M.Ventura

Ship Dimensioning

33

EXCEL Solver (2)

Multiple range
<Changing Cells> can be
indicated separated by
commas:
$C$4, $C$6:$C$8

<Target cell> is the one where the objective function is evaluated

<Equal To> define the type of problem (maximize, minimize, equal to)
<Changing Cells> are the ones that contain the design variables (to be
optimized) and must be all in the active sheet
<Constrains> list the constraints to be applied

M.Ventura

Ship Dimensioning

34

17

Example: Bulk Carrier Dimensioning

The simplified Model used on the example is based on the
one presented in Xuebin (2009)

Objective Function
M.Ventura

Ship Dimensioning

35

Example: Problem Constraints

The following set of 14 constraints is applied:

M.Ventura

Ship Dimensioning

36

18

Example: Bulk Carrier Dimensioning

M.Ventura

Ship Dimensioning

37

Notes on the Spreadsheet Design (1)

In order to make the model formulas more readable and
easy to debug and maintain, cell and range names should be
used instead of just references
Cell names are created by:
<Insert/Name/Define>
The use of cell names avoids the need to use absolute cell
references (Example: Lpp instead of C$4$)
Define all the cell names BEFORE entering the formulas
Define explicitly the units of all the values

M.Ventura

Ship Dimensioning

38

19

Notes on the Spreadsheet Design (2)

Constraints associated to intervals must be split in two.
Example:

25000 <= DW <= 500000

To be split into:
DW >= 25000
DW <= 50000

Use color codes to identify the different types of cells. For

example:
<yellow>

input cells

<orange> constraints
<red>

objective function

<gray>

values computed by the model

M.Ventura

Ship Dimensioning

39

Notes on the Spreadsheet Design (3)

Container Carriers have dimensions external (breadth of ship) and

internal (inner breadth of cargo hold) multiple of the width of the
standard container (8.0ft = 2.44m)
These conditions can be converted into additional constraints
For example for the Breadth of the ship:
Module(B/2.46) < 0.01

NOTES:
The value 2.46 results from taking into consideration the width of
the container plus the interval between containers (abt. 25 mm)
In Excel the expression will be:
mod(B; 2.46)
where the function mod(a;b) returns the remainder of the division
a/b
M.Ventura

Ship Dimensioning

40

20

Notes on the Spreadsheet Design (3)

Sometimes it is convenient to be able to use more than one

objective in the optimization process
Although the <Solver> is a single objective method, multiple
objectives can be taken into consideration by creating an objective
function which is the result of a weighted sum of several
contributions:
Fobj = w0xF0 + w1xF1 + w2xF2
The weights wi will be assigned by the designer in accordance to
the relative importance of each contribution and their sum will be
always equal to 1.0:
w0 + w1 + w2 + = 1.0
The sign of each weight will be positive, if the corresponding
contribution is to me minimized, or negative, if it is to be
maximized
It is convenient to scale the different contributions to the same
order of magnitude. For example each contribution can be scaled to
be in the interval [0, 1]

M.Ventura

Ship Dimensioning

41

Notes on the Spreadsheet Design (5)

The initial values for the variable cells should be
representative of the values expected at the optimal
solution, rather than arbitrary values such as all zeroes.
The Excel Solver is a local optimizer -> different sets of
initial variables values should be tested to check the
consistency of the results and to help to find a global
optimum
The process can be made automatic by creating a macro to
define the design variables, the constraints and the
objective function and to run the Solver
The macro can be converted into a VBA (Visual Basic for
Applications) function

M.Ventura

Ship Dimensioning

42

21

VBA Programming in Excel

The first draft of a
program can be obtained
by recording a sequence
of commands (macro)
using the macro
recorder:
<Tools/Macro/Record
New Macro>
Next the macro code can be run and edited in the VBA Editor
<Tools/Macro/Macros/Run> or /Edit>
The code should be extensively commented in order to make its
debugging and maintenance easier
M.Ventura

Ship Dimensioning

43

VBA Function for Dimensioning (1)

Sub OptimumShip()
'
' OptimumShip Macro
' Macro recorded 2010-09-22 by Manuel Ventura
'
' Keyboard Shortcut: Ctrl+Shift+S
'
' Clear Solver options
SolverReset
' Minimize Objective Function
SolverOk SetCell:="$K$17", MaxMinVal:=2, ValueOf:="0", _
ByChange:="$C$4:$C$9
' Constraints
' Lpp <=
SolverAdd CellRef:="$C$4", Relation:=1, FormulaText:="$C$24"

M.Ventura

Ship Dimensioning

44

22

VBA Function for Dimensioning (2)

' T <=
SolverAdd CellRef:="$C$7", Relation:=1, FormulaText:="$C$25
' T <=
SolverAdd CellRef:="$C$7", Relation:=1, FormulaText:="$C$26"
' L/B >=
SolverAdd CellRef:="$C$11", Relation:=3, FormulaText:="$C$27"
' L/D <=
SolverAdd CellRef:="$C$12", Relation:=1, FormulaText:="$C$28"
' L/T <=
SolverAdd CellRef:="$C$13", Relation:=1, FormulaText:="$C$29"
' Cb >=
SolverAdd CellRef:="$C$8", Relation:=3, FormulaText:="$C$30"
' Cb <=
SolverAdd CellRef:="$C$8", Relation:=1, FormulaText:="$C$31"
' Fn <=
SolverAdd CellRef:="$G$15", Relation:=1, FormulaText:="$C$32"

M.Ventura

Ship Dimensioning

45

VBA Function for Dimensioning (3)

' GMT >=
SolverAdd CellRef:="$G$31", Relation:=3, FormulaText:="$C$33"
' DW >=
SolverAdd CellRef:="$G$20", Relation:=3, FormulaText:="$C$34
' DW <=
SolverAdd CellRef:="$G$20", Relation:=1, FormulaText:="$C$35"
' Vs >=
SolverAdd CellRef:="$C$9", Relation:=3, FormulaText:="$C$36
' Vs <=
SolverAdd CellRef:="$C$9", Relation:=1, FormulaText:="$C$37"
' Run Solver and
allow the user to decide to keep or not the obtained result
SolverSolve userFinish:=False
End Sub
M.Ventura

Ship Dimensioning

46

23

Solver Options (1)

The Solver can be fine-tuned by changing the default
options

M.Ventura

Ship Dimensioning

47

Solver Options (2)

The <Max Time> and the <Iterations> edit boxes control the
Solvers running time.
The <Show Iteration Results> check box instructs the
Solver to pause after each major iteration and display the
current "trial solution" on the spreadsheet. In alternative
the user can simply press the ESC key at any time to
interrupt the Solver, inspect the current iterate, and decide
whether to continue or to stop.
The <Assume Linear Model> check box determines whether
the simplex method or the GRG2 nonlinear programming
algorithm will be used to solve the problem.
The <Use Automatic Scaling> check box causes the model to
be rescaled internally before solution.
M.Ventura

Ship Dimensioning

48

24

Solver Options (3)

The <Assume Non-Negative> check box places lower bounds
of zero on any decision variables that do not have explicit
bounds in the <Constraints> list box.
The <Precision> edit box is used by all of the optimizers and
indicates the tolerance within which constraints are
considered binding and variables are considered integral in
mixed integer programming (MIP) problems.
The <Tolerance> edit box is the integer optimality or MIPgap tolerance used in the branch and bound method.
The GRG2 algorithm uses the <Convergence> edit box and
<Estimates>, <Derivatives>, and <Search> option button
groups.

M.Ventura

Ship Dimensioning

49

Some Optimization Software Tools:

Matlab Optimization Toolbox

25

MatLab Optimization Toolbox fmincon()

M.Ventura

Ship Dimensioning

51

MatLab Optimization Toolbox fmincon()

M.Ventura

Ship Dimensioning

52

26

Objective Function

M.Ventura

Ship Dimensioning

53

Non-Linear Constraints

M.Ventura

Ship Dimensioning

54

27

Example: Bulk Carrier Dimensioning (1)

The simplified Model used on the example is based on the
one presented in Xuebin (2009)
This is a part of the Matlab code to call the optimizer:
% Initial point
Lpp = 185.0;
B = 26.0;
D = 14.5;
T = 10.5;
Vs = 15.0;
Cb = 0.70;
x0 = [Lpp B D T Vs Cb];
% Call optimizer
[x, acc, exitflag, output] = fmincon( @CalcModel, x0, [], [], ...
[], [], [], [], @mycon, options );
M.Ventura

Ship Dimensioning

55

Example: Bulk Carrier Dimensioning (2)

The file <CalcModel.m> defines the sequence of the
calculations required to compute the objective function:
function [annualCargoCost] = CalcModel( x )
% Design independent variables
Lpp
= x(1);
B
= x(2);
D
= x(3);
T
= x(4);
Vs
= x(5);
Cb
= x(6);
displ = 1.025*Lpp*B*T*Cb;
% Froude Number
Fn = 0.5144*Vs/sqrt(9.8065*Lpp);

M.Ventura

Dimensioning
annualCargoCostShip
= aoc/nvr/cdw;

56

28

Example: Bulk Carrier Dimensioning (3)

The file <mycon.m> contains the definition of the constraints:
function [c, ceq] = mycon( x )
global

Fn dw;

Lpp = x(1);
B = x(2);
D = x(3);
T = x(4);
Vs = x(5);
Cb = x(6);
% Stability
kb = 0.53*T;
bmt = (0.085*Cb - 0.002)*B*B/T/Cb;
kmt = kb + bmt;
gmt = kmt - (1.0 + 0.52*D);
M.Ventura

Ship Dimensioning

57

Example: Bulk Carrier Dimensioning (4)

The file <mycon.m> with the definition of the constraints:
% Inequality Constraints defined as
% ax + b <= 0
c = [-Lpp/B+6.0 Lpp/D-15.0 Lpp/T-19.0 ...
T-0.45*dw^0.31 T-0.7*D-0.7 ...
25000-dw dw-500000 ...
0.63-Cb Cb-0.75 ...
14.0-Vs Vs-18.0 ...
Lpp-274.32 Fn-0.32 ...
-gmt+0.07*B];
% NO equality constraints
ceq = [];

M.Ventura

Ship Dimensioning

58

29

Example: Bulk Carrier Dimensioning (4)

Final results of the optimization :
Algorithm used : medium-scale: SQP, Quasi-Newton, line-search
No. of iterations = 18
No. function calls = 133
Optimum Ship:
Lpp = 221.855 m
B = 36.976 m
D = 19.821 m
T = 14.575 m
Vs = 14.000 knots
Cb =
0.720
ACC =

7.972 US$/t

The results are quite similar to those obtained from the Excel
spreadsheet using the Solver.
M.Ventura

Ship Dimensioning

59

Linear Programming (LP) Methods Applied to

Ship Dimensioning

30

Introduction
Linear Programming (LP), is an Operations Research
technique that was first applied during the Second World
War to help solve troop-supply problems.

M.Ventura

Ship Dimensioning

61

LP Software Tools Available

MatLab Optimization Toolbox
Quadric and Linear Programming

LP Solve (ANSI C)
Current version: 5.5 (CD-ROM#68)

Clp - COIN-OR Linear Programming Solver (C++)

Current version: 1.10 (CD-ROM#68)

M.Ventura

Ship Dimensioning

62

31

Bibliography

Bibliography (1)
9 Artana, Ketut Buda and Ishida, Kenji (2003), "The Determination
of Optimum Ships Design and Power Prediction Using Spreadsheet
Model", Journal of the JIME Vol. 37, No. 6.
9 Artana, K.Buda and Ishida, Kenji (2003), "Spreadsheet Modeling to
Determine Optimum Ship Main Dimensions and Power Requirements
at Basic Design Stage", Marine Technology, Vol. 40, No. 1,
Jan.2003, pp. 6170. (CD-ROM#51)
9 Brinati, HL; Augusto, AO and Conti, MB (2007), Learning Aspects
of Procedures for Ship Concept Design Based on First Principles,
International Conference on Engineering Education ICEE2007,
Coimbra, Portugal.
9 Burgos, D. and Martins, M. (2008), "Projeto Preliminar de
Embarcaes Usando Algoritmos Genticos", SOBENA 2008.
9 Chao, Chen (2009), "The Container Shipping Network Design under
Changing Demand and Freight Rates", The Eighth International
Symposium on Operations Research and Its Applications
(ISORA09) Zhangjiajie, China, September 2022, pp. 245262.
(CD-ROM#68)
M.Ventura

Ship Dimensioning

64

32

Bibliography (2)
9 Chryssostomidis, Chryssostomos (1967), "Optimization Methods
Applied to Containership Design", MsC Thesis, MIT. (CD-ROM#67)
9 Cudina, Predrag (2008), "Design Procedure and Mathematical
Models in the Concept Design of Tankers and Bulk Carriers",
Brodogradnja, Vol.59, No.4, pp.323-339. (CD-ROM#70)
9 De, Abhijit and Kumar, Ashish (2006), "OPTI-MARINE-WARE
(Optimization of Vessel's Parameters Through Spreadsheet
Model)", Journal of Naval Architecture and Marine Engineering.
(CD-ROM#51)
9 Dobie, Thomas (2002), The Importance of the Human Factors in
Ship Design
9 Frank, Darko; Klanac, Alan and Bralic, Svemir (2008), "A Concept
for Concurrent Group Design of Ships", Proceedings of
COMPIT'08, Liege, pp.450-459. (CD-ROM#68)
M.Ventura

Ship Dimensioning

65

Bibliography (3)
9 Frank, D.; Klanac, A. and Bralic, S. (2008), "ng.zine - A New Design
System for Naval Architecture", Proceedings of SORTA'08, Pula.
(CD-ROM#68)
9 Ganesan, Vikram (2001), "Global Optimization of the Nonconvex
Containership Design Problem Using the Reformulation-linearization
Technique", MSc Thesis, Virginia Polytechnic Institute and State
University. (CD-ROM#67)
9 ISSC 2003, Technical Committee IV.2 Report

Jensen, G. (1994), Moderne Schiffslinien, Handbuch der Werften,

Vol.XXII, Hansa, pp.93.

9 Klanac, A. and Jelovica, J. (2007), "A Concept of Omni-Optimization

for Ship Structural Design", Advancements in Marine Structures,
Guedes Soares & Das (eds), Proceedings of MARSTRUCT 2007, The
1st International Conference on Marine Structures, 12-14 March
2007, Glasgow, UK. p. 473-481. (Taylor & Francis: London). (CDROM#68)
M.Ventura

Ship Dimensioning

66

33

Bibliography (4)
9 Liang, Zheng Xuan; Yan, Lin and Shang, Ji Zhuo (2009),
"Collaborative Multidisciplinary Decision Making Based on Game
Theory in Ship Preliminary Design", Journal of Marine Science and
Technology, Vol.1, pp.334344. (CD-ROM#70)
9 Murphy, R.; Sabat, D. and Taylor, R., Least Cost Ship
Characteristics by Computer Techniques (CD-ROM#33)
9 Na, S-S and Karr, D. (2002), Product-Oriented Optimal Structural
Design of Double-Hull Oil Tankers, Journal of Ship Production, Vol.
18, No. 4, Nov. 2002, pp. 237248. (CD-ROM#51)
9 Parsons and Scott (2004), Formulation of Design Optimization
Problems for Solution with Scalar Numerical Optimization
Methods, Journal of Ship Research, Vol.48, No.1, pp.61-76. (CDROM#51)

M.Ventura

Ship Dimensioning

67

Bibliography (5)
9 Peri, D. and Campana, E. F. (2003), Multidisciplinary Design
Optimization of a Naval Surface Combatant, Journal of Ship
Research, Vol.47, No.1, pp.1-12. (CD-ROM#51)
9 Ross, J.; McNatt, T. and Hazen, G. (2002), The Project 21 Smart
Product Model: A New Paradigm for Ship Design, Cost Estimation,
and Production Planning, Journal of Ship Production, Vol. 18, No. 2,
May 2002, pp. 7378. (CD-ROM#33)
9 Schiller, T.R.; Daidola, J. C.; Kloetzli, J.C. and Pfister, J. (2001),
"Portfolio of Ship Designs: Early-Stage Design Tools", Marine
Technology, Vol.38, No.2, April 2001, pp.7191. (CD-ROM#51)
9 Schneekluth, H. e Bertram, V. (1998), Ship Design for Efficiency
and Economy, Butterworth-Heinemann.
Watson, DGM and Gilfillan, AW (1976), Some Ship Design
Methods, RINA Transactions, Vol.119, pp.279-324.
Whiton, Justin C. (1967), "Some Constraints On Shipping in Linear
Programming Models", Naval Research Logistics Quarterly, Vol. 14,
Issue 2, pp.257-260.
M.Ventura

Ship Dimensioning

68

34

Bibliography (6)
9 Xinlian, Xie; Tengfei, Wang and Daisong, Chen (2000), "A Dynamic
Model and Algorithm for Fleet Planning", Maritime Policy &
Management, Vol.27, Issue 1, pp.53-63. (CD-ROM#68)
9 Xuebin, Li (2009), Multiobjective Optimization and Multiattribute
Decision Making Study of Ships Principal Parameters in Conceptual
Design, Journal of Ship Research, Vol.53, No.2, pp.83-92.
9 Yang, Y-S; Park, C-K; Lee, K-H and Suh, J-C (2007), A Study on
the Preliminary Ship Design Method Using Deterministic and
Probabilistic Approach Including Hull Form, Journal of Structural
Multidisciplinary Optimization, Vol.33, No.6, pp.529-539. (CDROM#65)
9 Zanic, Vedran and Cudina, Predrag (2009), "Multiattribute Decision
Making Nethodology in the Concept Design of Tankers and Bulk
Carriers", Brodogradnja, Vol.60, No.1, pp.19-43. (CD-ROM#70)

M.Ventura

Ship Dimensioning

69

Bibliography
Linear Programming
9 Ferris, Michael C.; Mangasarian, Olvi L. and Wright, Stephen J.
(2007), Linear Programming with MatLab", Society for Industrial
and Applied Mathematics and the Mathematical Programming
Society.
9 Luenberger, D.G. anf Ye, Y. (2008), Linear and Non-Linear
Programming , 3rdEd, Springer.
9 Matousek, Jiri and Gartner, Bernd (2006), "Understanding and
Using Linear Programming", Springer.
Linear Programming Applied to Ship Design

Moyst, Howard and Das, Biman (2008), A Linear Programming

Approach to Optimization of Ship Design and Construction Phases,
Journal of Ship Production, Vol. 24, No.1, pp. 1-6.

M.Ventura

Ship Dimensioning

70

35

Annex A. Ships Statistical Data Gathering

and Processing

Ships Data Gathering (1)

The common practice of single design or small series implies that

some initial knowledge can be obtained from the analysis of the
existing ships

To improve the efficiency of the process the information about

existing ships of the same type and in a similar range of cargo
capacity should be structured in a small Data Base

To improve the quality of the process, the Data Base should first
be cleaned from:

Incorrect data (from wrong sources or typing mistakes)

Incomplete data (incomplete records with some missing fields)

Repeated data (from identical ships produced in series)

Keeping track of the ship identification (Name, IMO Number,

building yard and year) and data source (journal, web site, etc.)
will help to check and improve the data quality

M.Ventura

Ship Dimensioning

72

36

Ships Data Gathering (2)

The registries of Lloyds Register and other classification
societies are good data sources
A Spreadsheet can be used for data storage and for the
statistical analysis and graphic display of the results
The main topics of interest are:
Hull dimensions
Propulsion machinery and electric generators
Cargo capacity and equipment
Others (ballast capacity, crew)
Name

Built
Year

IMO
No

M.Ventura

Lpp

Lwt

DW

Vs

Main
Engine

MCR

Ship Dimensioning

Electr.
Power

73

Ships Data Gathering (3)

The measure(s) of the cargo capacity used depends of the ship type:
Ship Type

Measures of Cargo Capacity

Cargo
Equipment

Tankers
Bulk carriers

No. of Tanks / Holds

Volume cargo tanks/holds

Cargo pumps
Cranes

Container carriers

Total Number of TEUs (in holds,

on deck, reefers)

Cranes
Cell guides

Ro/Ro
RoPax
Ferries

Total lane length / number of cars

/ number of trailers
Number of passengers

Ramps, lifts

Passenger Ships

M.Ventura

Number of passengers

Ship Dimensioning

74

37

Ships Data Gathering (4)

Other information also useful about ship systems, crew, etc.
(if available)
Vol.
Ballast

Ballast
Pumps

M.Ventura

Vol. FO

Vol. DO

Crew

Ship Dimensioning

75

Process the Compiled Data

Based on the data compiled, a set of ratios can be computed
These ratios help to characterize the ship class
Allow the definition of the bounding limits to the variation
of the design variables
Support the estimative of values for which there is no
information to support a computation, even if approximated
L/B

B/T L/D T/D

M.Ventura

Cb

LWT/(L.B.D)

CSR WB/DW

Ship Dimensioning

%TEUdeck
%TEU hold

%TEUref

76

38

Ship Data Sources Web Sites (1)

Data Source

Notes

exchange.dnv.com

DNV Registry of ships and characteristics. Free access.

Search by Name or IMO Number

www.sea-web.com

Lloyds Register Data about existing ships (paid)

polship.cto.gda.pl

Data about ships built in Polish shipyards since 1992

M.Ventura

Ship Dimensioning

77

Ship Data Sources Publications (2)

Data Source

Notes

Motor Ship
www.motorship.com

Journal that contains some ship descriptions

Digital version available only by subscription.

Ingenieria Naval
Journal of the association of Spanish naval architects that
www.ingenierosnavales contains some ship descriptions (in Spanish). Digital version
available only by subscription.
.com
Naval Architecture

RINA Journal that contains some ship descriptions

Significant Ships

Annual publication from RINA with good descriptions of the

most representative ships of each year, including General
Arrangement drawing and lightship weight information. Also
available in CD-ROM.

M.Ventura

Ship Dimensioning

78

39

Annex B. Physical Limitations to the Main

Dimensions of the Ship

Physical Limitations
Physical limitations can be associated to the geographical
route that the ship uses
Limitations can be due to the existence of canals, straights,
bridges, ports, locks systems
The dimensions affected can be the Length, the Breadth,
the Draught and the Air Draught

M.Ventura

Ship Dimensioning

80

40

Air Draught
Designation given to the
vertical distance measured
from the load waterline up
to the upper extremity of
the ship (top of the mast,
chimney,..)
Limited to 4.50 m in many
inland waterways in central
Europe due to the
existence of bridges

M.Ventura

Ship Dimensioning

81

Some Physical Limitations in Canals

Lmax
[m]

Bmax
[m]

Tmax
[m]

Air
Draft
max [m]

DW Max.
[t]

TEU Max.

Panama Canal

294.13

32.31

12.04

--

65,000

4,000

Panama Canal
(after 2014)

427.00

55.00

18.00

--

Kiel Canal

315.00

40.00

9.50

--

St. Lawrence
Canal

222.50

22.86

9.10

35.50

Suez Canal

---

--

20.12

--

240,000

17,000

Strait of Malacca

---

---

21.00

--

300,000

18,000

M.Ventura

Ship Dimensioning
Updated
on Jan. 2010

12,000

82

41

Strategic Points for the Marine

Transportation

M.Ventura

Ship Dimensioning

83

Panama Canal

M.Ventura

Ship Dimensioning

84

42

Enlargement of the Panama Canal

Adapted to Post-Panamax ships
Dimensions of the new locks (eclusas):
L = 427 m
B = 55 m
T = 18 m
Cost: 5.5 billion US$
Beginning of work: 2007
Conclusion:

M.Ventura

2014

Ship Dimensioning

85

Suez Canal Navigation Chanel

M.Ventura

Ship Dimensioning
Updated
on Jan. 2010

86

43

Suez Canal Evolution of the Cross

Section Dimensions

M.Ventura

Ship Dimensioning

87

Strait of Malacca

M.Ventura

Ship Dimensioning

88

44

Restrictions in Portuguese Ports

Length
Leixes/
tankers

Breadth

Station B = 200 m
Station C = 100 m

Leixes/
Other ships

Draught

Air Draught

Station B = 9 m
Station C = 5.8 m
?

Aveiro

140 m

8m

Figueira da Foz

100 m

4.7 m

Lisboa

Trafaria = 235 m

Barra= 10.5 m
(dep. on the tide)
Liscont = 10 m
Sta.Apol. = 8 m
Trafaria = 12 m
Barreiro = 9 m
Seixal = 5m

Setbal

250 m

Barra= 9.5 m
(dep. on the tide)

Sines
M.Ventura

Gen.Cargo = 125 m

Petrol.= 28 m
Grain.= 17 m
Gen.Cargo= 5.5 m
Ship Dimensioning

Petrol.= 23 m
Grain.= ?

89

Marine Ports in Portuguese Coast

M.Ventura

Viana do Castelo

Leixes

Aveiro

Figueira da Foz

Peniche

Lisboa

Cascais

Sesimbra

Setbal

Sines

Lagos

Faro

V. R. Sto. Antnio

Ship Dimensioning

90

45

Links Portuguese Ports

www.portosdeportugal.pt
www.portodeaveiro.pt
www.portodelisboa.pt
www.portodelisboa.pt
www.portodesines.pt
www.portodesetubal.pt
www.apdl.pt

M.Ventura

Ship Dimensioning

91

Links Ports and Canals

www.pancanal.com

(Panama Canal)

www.suezcanal.gov.eg

(Suez Canal Authority)

www.kiel-canal.org

(Kiel Canal)

www.greatlakes-seaway.com
www.atlas.com.eg/scg.html
www.nnc.egnet.net/suezrules.htm
www.portguide.com

M.Ventura

Ship Dimensioning

92

46

Annex C. Economical Measures of Merit

Measures of Merit
The type of measure of merit used depends on the previous
knowledge of the earnings of the ship
Known Results
Net Present Value (NPV)
Internal Rate of Return (IRR)

Unknown Results
Required Freight Rate (RFR)
Present Value (PV)
Average Annual Cost (AAC)

M.Ventura

Ship Dimensioning

94

47

Net Present Value (NPV)

Often used when the funds for investment are limited and
the maximum income tax possible is required.

NPV =

P W

(Q

FR ) PW

(AOC )

PW

( C 0 )

where:

- No. years of ships life

PW() - Present Worth
Q
- Total quantity of cargo carried annually
FR
- Freight Tax
AOC Annual Operating Costs
C0
- Initial ship cost

M.Ventura

Ship Dimensioning

95

Internal Rate of Return (IRR)

Represents the tax of return which originates equal values
for the Present Value of the results and of the costs, i.e.,
for which NPV = 0.
Allows more effective comparisons between entirely
different alternatives
While NPV is expressed in currency units (Euro, US$), the
IRR is expressed in percentage (%)
One advantage of the IRR is that it can be computed without
the need to estimate the cost of the capital
When the IRR is used, the criterion is to select the projects
whose IRR exceeds the cost of the capital

M.Ventura

Ship Dimensioning

96

48

Required Freight Rate (RFR)

Used when the data necessary to determine accurately the

exploitation results is not available.

Specially advantageous when comparisons are made between

ships of different sizes.

Represents the cost per unit of cargo, necessary to cover

entirely the operation costs and to guarantee the specified
income tax from the capital invested.

RFR =

AOC + Ci
Q

where Ci is the annual cost of the capital and VR is the

residual value of the ship

Ci = CRF ( C0 PW VR )

M.Ventura

CRF = Capital Recovery

Factor

Ship Dimensioning

97

Permissible Price (PP)

Represents the maximum admissible price of the ship that

still guarantees a specified income tax.

With the exception of the cases where the ship is paid in a

single installment, it is determined by an iterative
process.

Can be used to evaluate prices in proposals of new buildings

or in the acquisition of second-hand ships, and in the
comparison of those prices with the current ship prices and
freight values.

M.Ventura

Ship Dimensioning

98

49

Some Elementary Concepts

Capital Recovery Factor - is factor that converts a present
value into a stream of equal annual payments over a specified
period of time at a given interest rate.

CRF =

i (1 + i )

(1 + i )

where:
N No. years of ships life
i
Interest rate

Present Worth Factor - is a multiplier which converts a

future amount into a present amount

PW = (1 + i )
M.Ventura

Ship Dimensioning

99

Bibliography
9 BTE (1982), An Estimate of Operating Costs for Bulk, RoRo and
Containers Ships, Bureau of Transport Economics, Camberra.
9 Watson, D.G.M. (1998), Practical Ship Design, Vol.1, Elsevier.
9 Y-S Yang, C-K Park, K-H Lee and J-C Suh (2007), A Study on the
Preliminary Ship Design Method Using Deterministic Approach and
Probabilistic Approach Including Hull Form, Structural and
Multidisciplinay Optimization, Vol.33, No.6, pp.529-539. (CDROM#50)

M.Ventura

Ship Dimensioning

100

50

Some Relevant Links (1)

Web Site

Description / Notes

www.awes-shipbuilding.org

AWES Association of European Shipbuilders and

Shiprepairers

www.cesa-shipbuilding.org

CESA Committee of European Shipyards

Associations

www.dataloy.com

Routes and distances between ports

www.priyablue.com

Prices of second-hand ships

www.bunkerworld.com

Updated prices of the different types of Fuel Oil.

Historic record and trends.

www.brs-paris.com

BRS Barry Rogliano Salles Shipbrokers statistics

about the marine transport market

www.equasys.org

Ship search by IMO Number

M.Ventura

Ship Dimensioning

101

Some Relevant Links (2)

Web Site

Description / Notes

e-ships.net

World Shipping Register (subscription)

Search by Name, Type, DW, etc.

www.shippingdatabase.com

Ship search by IMO Number, Name, Owner (free

registration)

M.Ventura

Ship Dimensioning

102

51

Some Portuguese Links

www.imarpor.pt (Instituto Porturio dos Transportes Martimos)

www.ancruzeiros.pt (Lista de Legislao Nutica de Recreio)

www.fpvela.pt (Federao Portuguesa de Vela)

www.hidrografico.pt (Instituto Hidrogrfico)

www.isn.org.pt (Instituto de Socorros a Nufragos)

M.Ventura

Ship Dimensioning

103

52