This paper presents a method for optimizing statistical multi-response problems. The method
is based on fuzzy goal programming and it enjoys a strong mathematical foundation. In this
method, the decision maker's comments are considered objectively. The LINGO programming
environment is used to test the developed method. The method performance is evaluated by
comparing the results with those of other existing methods.

In many countries, there are traditional houses made of stone, adobe and brick, which perform
relatively weakly in earthquakes. To mitigate casualties in future earthquakes, it is necessary to
evaluate the current status of these buildings and propose eective methods to retrot them. One
method of evaluating the performance of buildings in earthquakes is through the use of fragility
curves. These diagrams show the probability of exceeding a specic state of damage versus
seismic intensity parameters, such as PGA, MMI, Ia, CAV etc. Fragility curves will provide
an important basis for analytical methods, based on probabilistic approaches. Much work has
been done on bridges, concrete and steel structures, while limited studies have been conducted
on masonry buildings. Considering the prevalence of these buildings and the high seismic activity
in some regions where they are built, fragility curves for dierent types of masonry buildings are
developed in this research. The results obtained show the probable damage to those types of
masonry building chosen in this research for dierent earthquake intensities and restates that, by
providing horizontal and vertical ties and retrotting these buildings, by the methods proposed
in some seismic code provisions (such as FEMA 356 and 357), their behavior in earthquakes are
apparently eected and structural damage reduced. Furthermore, results indicate that soil type
does not signicantly in
uence the seismic behavior of masonry buildings.

A sensitivity based parameter identication method is presented to detect the damage of existing
structures, using applied sets of static forces at one subset of degrees of freedom and measured
displacements at a subset of degrees of freedom that may overlap completely, partially, or not
at all. The algorithm follows an output error approach, which minimizes the deviation between
a measured and a theoretical displacement, in lieu of the commonly used force error function.
An iterative scheme is developed utilizing a rst order Taylor series expansion to linearize the
associated non linear problem. The algorithm automatically adjusts the structural element
stiness parameters, in order to improve the comparison between a measured and a theoretical
response in an optimal way. The measured input required in the present study is articially
generated. The eect that a noisy displacement measurement has on an identication procedure
is also studied. A procedure is also identied, in order to select the limited number of DOF
required to perform successful parameter identication, reducing the impact of measurement
errors on the identied parameters. The algorithm is elucidated by a numerical example on
frame structures.

In this paper, a procedure is proposed for the evaluation of displacement gradients in a twodimensional,
cell centred, nite volume formulation for stress analysis in linear elastic solid
problems. Temporary elements with isoparametric formulations are used for calculation of the
gradients at the cell boundaries. In this way, stress continuity across the common face of the
two adjacent cells will be guaranteed. The formulation is veried by three test cases, in which
the proposed formulation shows good predictions.

In this paper, the eect of geometrical parameters on the fretting fatigue of a half-plane in contact
with a
at and rounded pad is studied. This is accomplished by calculating and comparing the
stress states and stress intensity factors of fretting cracks for a number of pad geometries. The
pad geometry is represented by the radius of its rounded corners and the width of its central
at
part. The distribution of dislocation method is employed to calculate the stress intensity factors
of fretting induced cracks of dierent lengths for dierent values of geometrical parameters. The
results of this study can be insightful for improving the geometrical design of an aero-engine
compressor disk, as similar contact and damage are prone to occur in its dovetail region.

The present study is concerned with the numerical simulation of free-surface waves and wave
induced separation in the presence of an intrusion. The results of several simulations are reported.
The rst study was performed for a NACA0024 surface piercing hydrofoil over a range of several
Froude numbers (0.19, 0.37, 0.55), along with wave breaking at Fr = 1.0 The NACA0024 foil was
of particular interest, as it almost has no separation at large depths

thus the eect of the freesurface
wave and the wave induced separation could be studied. Free- surface waves and wave
induced separation results were evaluated and compared with both the available experimental
data and the previous numerical results. The wave breaking
ow was also successfully simulated
and results were presented. The second series of simulations were carried out for a circular cylinder
in order to investigate the shape eects on the wave-induced separation. Results suggest that,
at high Froude numbers, the free-surface waves are a function of Froude and not the shape of a
body. Flow features with regard to separation, free-surface elevations and drag coecients were
also studied.

In the boundary element formulation of heat conduction, the heat source eect imposes an
additional domain integral term on the system of integral equations. With this term, an
important advantage of the boundary element method as a boundary-only formulation will
be lost. This paper presents an accurate method for the evaluation of heat source domain
integrals, with no need of domain discretization. Transformation of the domain integral into the
corresponding boundary integral is carried out using Green's theorem. Both time-dependent and
time-independent fundamental solutions are considered. The methodology can be implemented
in general and for similar situations. Numerical examples will be presented to demonstrate the
accuracy and eciency of the presented method.

The analysis and simulation of the manufacturing process require extensive and complicated
computations. Nowadays, computer resources and computational algorithms have reached the
stage where they can model and simulate the problem eciently. One of the important processes
in manufacturing is machining. In this research, the end-milling process, which is one of the most
complex and widespread processes in machining, is chosen. The most important parameters in
end-milling are surface roughness and surface location errors. Comprehensive simulation software
is developed to model the end-milling process, in order to anticipate the nishing parameters, such
as surface roughness and errors. The proposed algorithm takes into account cutting conditions,
such as feed, doc, woc and tool run out etc. In addition, the dynamic simulation module
of the software can accurately model the
exible end-mill tool, the milling cutting forces and
regeneration of the waviness eects, in order to construct a realistic surface texture model.
The software can accurately determine the most commonly used index of surface roughness
parameters, such as Ra, P.T.V. and surface errors.

In this paper, the need to incorporate three famous design cost management methods, called:
Quality Function Deployment (QFD), Value Engineering (VE) and Target Costing (TC) into
a single model has been addressed. Each method performs very well in cost management
procedures as design activities. These methods have been incorporated into a mathematical
programming model, in order to achieve the maximum benet of each method. The model,
essentially, optimizes customer satisfaction subject to target cost. The tool is a mixed integer
zero-one nonlinear programming. The unied model has been proposed to prevent a non-optimal
solution when methods interact with each other. The practitioner should be condent that the
quality solution would be achieved in contrast to when the methods are applied sequentially. A
simple automobile design example was formulated and solved to show the performance of the
model.

Thee enclosed basin has certain natural frequencies of seiche, depending on the geometry of
the water boundaries and the bathymetry of water depths. Therefore, the variation in the water
surface at a point becomes irregular, as caused by the combination of several natural frequencies,
which may be considered as the superposition of sinusoidal frequency components of dierent
amplitude. This paper is mainly concerned with the motion of an incompressible irrotational
uid
in a closed rectangular basin with internal impervious barriers. An analytical solution is presented
for predicting the characteristic of generated waves in these types of basin. The equations of free
water surface oscillations and its boundary conditions are reduced to a system of linear equations,
which is solved by applying the small amplitude water wave theory. The
ow potential, wave
amplitude,
ow patterns and the natural period of waves generated in the basin with impervious
internal barriers are found, based on the basin geometry. It is shown that the natural period of
the basin is strongly dependent on the location of the barriers and the size of the barrier opening.

In this paper, the large amount of highway runo characterization data that were collected in
California, during a 3-year monitoring season (2000-2003), were assessed in order to develop an
Articial Neural Network (ANN) model for predicting the Event Mean Concentration (EMC)
of the constituent. The initial data analysis performed by a Multiple Linear Regression (MLR)
model revealed that the Total Event Rainfall (TER), the Cumulative Seasonal Rainfall (CSR), the
Antecedent Dry Period (ADP), the contributing Drainage Area (DA) and the Annual Average
Daily Trac (AADT) were among the variables having a signicant impact on the highway
runo constituent EMC. These parameters were used as the basis for developing an Articial
Neural Network (ANN) model. The ANN model was also used to evaluate the impact of various
site and storm event variables on highway runo constituents' EMCs. The ANN model has
proven to be superior to the previously developed MLR model, with an improved R2 for most
constituents. Through the ANN model, one was able to see some non-linear eects of multi
variables on pollutant concentration that, otherwise, would not have been possible with a typical
MLR model. For example, the results showed that copper EMC is more sensitive at higher
Annual Average Daily Trac (AADT), with respect to ADP, compared with lower range AADT.

This paper presents the study of the structural and optical properties of copper doped cadmium
sulde thin lms prepared by pulsed Nd:YAG laser. The copper concentration in the targets was
varied from 0.05% to 5% in weight and they were heated at a temperature of 500C in air for
one hour. It was observed that annealing un-doped and doped thin lms at temperatures above
160C leads to an abrupt increase in the optical transmission, changing from dark brownish to
transparent yellowish. For annealed samples at 300C, increasing Cu concentration from 0.0%
to 2%, leads to an increase in the optical transmission up to 80%, at a wavelength of about 700
nm. XRD patterns showed the hexagonal phase of cadmium sulde and a number of CdO peaks
relating to the partial oxidation of cadmium. The XPS spectra of the annealed samples indicate
an accumulation of Cu at the surface. Therefore, Cu impurity can enhance CdS oxidation and
the size of the crystallites. The highest value of the photo-current for the samples is related to
the lowest Cu concentration, i.e: 0.05%.

In hierarchical service networks, facilities at dierent levels provide dierent types of service. For
example, in health care systems, general centers provide low-level services, such as primary health
care, while specialized hospitals provide high-level services. Because of the demand congestion
at service networks, the location of servers and their allocation of demand nodes can have
a strong impact on the length of the queue at each server, as well as on the response time to
service calls. This study attempts to develop hierarchical location-allocation models for congested
systems by employing a queueing theory in a fuzzy framework. The parameters of each model
are approximately evaluated and stated as fuzzy-numbers. The coverage of demand nodes is
also considered in an approximate manner and is stated by the degree of membership. Using
queueing theory and fuzzy conditions, both referral and nested hierarchical models are developed
for the Location Set Covering Problem (LSCP). To demonstrate the performance of the proposed
models, a numerical example is solved in order to compare the results obtained from the existing
probabilistic models and the new fuzzy models developed in this paper.

In this paper, a three-dimensional steady state incompressible turbulent air
ow is considered in
a large single room. The buoyancy aected turbulent air
ow is numerically simulated by solving
governing equations. The turbulence modeling includes both k " and zero-equation models
and their results are compared to the experimental data. The paper reviews several aspects,
such as displacement of radiator system performance, temperature and
ow eld distribution
and comfort conditions. The results show that the best temperature distribution and comfort
condition are obtained when the radiator is installed under the window and its height is equal to
or greater than that of the window.

The unaxisymmetric heat transfer of an unsteady viscous
ow, in the vicinity of an axisymmetric
stagnation-point of an innite circular cylinder, with simultaneous axial and rotational movement,
along with transpiration, Uo, is investigated, when the angular velocity, axial velocity and wall
temperature or wall heat
ux vary arbitrarily with time. The impinging free stream is steady and
with a strain rate of k. An exact solution of the Navier-Stokes equations and energy equation
is derived in this problem. A reduction of these equations is obtained by the use of appropriate
transformations for the most general case, when the transpiration rate is also time-dependent.
However, results are presented only for uniform values of this quantity. The general self-similar
solution of unsteady unaxisymmetric heat transfer is obtained, in which unaxisymmetry is due to
the sinusoidal variation of the temperature, with respect to the surface position of the rotating
cylinder, and unsteadiness is because of the sinusoidal variation of the temperature of each
point of the cylinder surface, with respect to time and, also the rotation of the cylinder. All
the solutions are presented for Reynolds numbers, Re = ka2=2, ranging from 0.1 to 10.0 for
dierent values of Prandtl number and selected values of the dimensionless transpiration rate,
S = U0=ka, where a is the cylinder radius and is the kinematic viscosity of the
uid. The
local coecient of heat transfer (Nusselt number) is found to be independent of time and place,
though the cylinder wall temperature or wall heat
ux are both functions of time and place.

Nowadays, various types of passive control systems are being used as an eective solution
to reduce the seismic responses of structures. One type of these systems, the Tuned Liquid
Column Damper (TLCD), suppresses the input seismic energy by a combined action, including
the movement of liquid mass in the container, a restoring force on the liquid, due to gravity
loads, and the damping, due to liquid movement through the orices. In this paper, the possible
eects of seismic excitation characteristics, such as frequency content and soil condition, on
the seismic performance of TLCDs, are investigated, using nonlinear time-history analyses.
For this purpose, a ten-story building was modeled as an elastic MDOF structure and used
for numerical analyses. For the time-history analyses, among the past strong ground motion
records of Iran, 16 records with dierent characteristics were selected. The results of this study
show that these characteristics play a substantial role in the performance of TLCDs and they
should be, accordingly, considered in the designing of TLCDs. In some cases, TLCD is able
to reduce structural displacement up to 50%, while, in most cases, the eectiveness of TLCD
in reducing structural acceleration is not signicant. However, it should be mentioned that, in
real applications, de-tuning may occur, due to the inelastic behavior of structures, which can
reduce eectiveness. This study also shows that the displacement reduction capacity of TLCDs
is highly dependent on excitation characteristics, while the acceleration reduction capacity is not
that sensitive.