EJSR(5)

European Journal of Scientific Research
ISSN 1450-216X / 1450-202X Vol.121 No.3, 2014, pp.310-320
http://www.europeanjournalofscientificresearch.com
A Hybrid SS-SA Approach for Solving Multi-Objective
Optimization Problems
Ramadan A. ZeinEldin
Operations Research department, Institute of Statistical
Studies ad Research, Cairo University
E-mail: rzeanedean@yahoo.com
Abstract
Decision makers, nowadays, face complex real world problems having more than
one conflicting objective functions to be optimized at the same time. In this paper, we
developed a hybrid approach based on scatter search and simulated annealing for solving
the multi-objective optimization problems. To validate our approach, we solved some test
problems from the literature, compared the results with other approaches, and found that
our proposed approach performs well.
Keywords: Multi-objective, optimization, Scatter search, simulated annealing.
1. Introduction
Multi-objective optimization is an important research topic for scientists and researchers. This is due to
the multi-objective nature of real world problems. Most real problems are complex and
multidisciplinary in nature, and quite often require more than one conflicting objective functions to be
optimized simultaneously while usually no prior information of their exact interactions is available.
Researchers have developed many multi-objective optimization procedures. For multi-objective
optimization problems, there is not a single optimum solution, but a set of non-dominated optimal
solutions called the Pareto set of solutions. The challenge is in the case of conflicting objectives, which
is usually the case in most real problems.
Traditional mathematical programming techniques have some limitations when solving MOPs.
Most of them depend on the shape of the Pareto front and only generate one Pareto solution from each
run. Thus, several runs (with different parameter settings) are generally required to generate a Pareto
solution set; however, sometimes different parameter settings may generate similar results. In such
circumstances, generating a Pareto solution set will be very computationally expensive [Li and Du,
2013]. In the last two decades most of the researchers used meta-heuristics approaches such as genetic
algorithm, simulated annealing, tabu search, particle swarm to solve the multi-objective problems.
Simulated annealing algorithm repeats an iterative neighbor generation procedure and follows
search directions that improve the objective function value. While exploring solution space, the SA
method offers the possibility of accepting worse neighbor solutions in a controlled manner in order to
escape from local minima.
In terms of metaheuristics, recently, scatter search approaches are receiving increasing
attention, because of their potential to effectively explore a wide range of complex optimization
problems [Silva et al. 2013]. Hybrid algorithm can make good use of the characteristics of different
algorithms to achieve complementary advantages to improve the algorithm optimal performance and
efficiency [Zhang et al. 2012].

A Hybrid SS-SA Approach for Solving Multi-Objective Optimization Problems 311
In this paper, a hybrid approach based on scatter search and simulated annealing is developed to
solve the multi-objective problems. The remainder of this article is organized as follows. The problem
definition and the literature review for the multi-objective optimization problem are presented in
section 2. Simulated annealing is presented in section 3 and scatter search is presented in section 4. The
proposed hybrid approach is presented in section 5, and section 6 is assigned to the implementation of the
proposed approach. Finally, the conclusions and points for future research are mentioned in section 7.
2. Multi Objectives and Literature Review
Multi-objective optimization problems contain more than one objective that needs to be achieved
simultaneously. Such problems arise in many applications, where two or more, sometimes competing
and/or incommensurable objective functions have to be minimized concurrently. This group of
optimization problems has a rather different perspective compared to single objective problems. In the
single objective optimization there is only one global optimum, but in multi-objective optimization
there is a set of solutions, called the Pareto-optimal set, which are considered to be equally important;
all of them constitute global optimum solutions (Kaveh and Laknejadi, 2011). Multi-objective
optimization requires more computational effort than single-objective optimization. Unless preferences
are irrelevant or completely understood, solution of several single objective problems may be
necessary to obtain an acceptable final solution.
The Multi-Objective Optimization Problem is a vector optimization problem stated as follows:
(Marler and Arora, 2004; Kaveh and Laknejadi, 2011; Bajestani et al., 2009)
Minimize: )}(.......)(),({)( 21 xf,xfxfXF p
Subject to: m,j,xg j 10)(  inequality constraints
l,k,xhk 10)(  equality constraints
n,i,xxx u
ii
l
i 1 side constraints
Most multi-objective-programming techniques focus on finding the set of efficient points for a
given problem or, in the case of heuristic procedures, an approximation of the efficient set (Molina et
al. 2007). The two critical challenges in MOP are: 1) The generation of a Pareto set of solutions; and 2)
the selection of a preferred final design from the Pareto set of solutions generated. Multi-objective
optimization solution strategies have been classified into the following categories, based on how
selecting a preference setting is implemented: 1) no preference method; 2) the a priori method, – where
a decision maker's preferences are taken into consideration before the optimization; 3) the a posteriori
method, – where a design is selected based on preferences after the Pareto set is generated; and 4)
interactive methods, – where the decision maker can incorporate and change preferences during the
solution process (Miettinen, 1999). In the last decade, evolutionary approaches have been the primary
tools to solve real-world multi-objective problems (Konak et al., 2006).
Marler and Arora (2004) presented a survey of predominant, continuous, and nonlinear multi-
objective optimization methods. Beausoleil (2006) presented an adaptation of the tabu/scatter Search
methods to multi-objective optimization. Bandyopadhyay (2008) developed a simulated annealing
based multi-objective optimization algorithm. Konak et al., (2006) presented multi-objective GA.
Talaslioglu (2010) proposed optimization tool based on scatter search methodology for the design of
grillage systems. Kaveh, and Laknejadi (2011) developed hybrid method combining particle swarm
method and charge system search. Mousa et al. (2012) introduced a hybrid multi-objective
evolutionary algorithm which combines genetic algorithms and particle swarm optimization. Molina et
al. (2007) developed a meta-heuristic based on scatter search and tabu search method for solving
nonlinear multi-objective optimization problems. Ho et al. (2003) proposed a simulated annealing
algorithm for multi-objective optimizations to find the Pareto solutions in a relatively simple manner.
Li and Du (2013) proposed a hybrid genetic algorithm to solve constrained multi- objective
optimization problems. Zou et al. (2013) proposed a teaching-learning-based optimization algorithm

312 Ramadan A. ZeinEldin
based on the nondominated sorting and crowding distance sorting for solving MOPs. Yang et al.
(2013) used the flower pollination algorithm to solve multi-objective optimization. Enayatifar et al.
(2013) developed a novel multi-objective evolutionary algorithm based on imperialist competitive
algorithm. Gong et al. (2013) presented a novel evolutionary algorithm that interacts with a decision
maker during the optimization process to obtain the most preferred solution. Jiao et al. (2013) proposed
a novel co-evolutionary multi-objective optimization algorithm based on direction vectors. Rahimi-
Vahed et al. (2007) introduced a multi-objective scatter search (MOSS) approach for the mixed-model
assembly line sequencing problem. Bajestani et al. (2009) presented a new MOSS for solving the
dynamic cell formation problem. Moghaddam et al. (2010) proposed a novel multi-objective scatter
search algorithm for location-routing problem. Silva et al. (2006) applied the scatter search
methodology for large size bi-criteria knapsack instances. Beausoleil (2006) presented a new improved
MOSS that employs a MultiStart TS approach as generator of a diverse set of initial solutions
permitting to obtain non-dominated solutions in promising regions.
3. An Overview about Simulated Annealing
Simulated annealing (SA) is a robust technique, which provides excellent solutions to single and
multiple objective optimization problems with a substantial reduction in computation time. Suman and
Kumar (2006) developed a comprehensive review of simulated annealing based optimization
algorithms for solving single and multi-objective optimization problems. SA has shown to be effective
in finding the optimal solution for both discrete and continuous optimization problems (Xhafa et al.,
2011).
SA is used to solve multi-objective optimization problems in different areas. Ekbal and Saha
(2011) proposed a simulated annealing based multi-objective optimization approach for classifier
ensemble. Jarosław et al. (2013) used SA algorithm to solve flow shop scheduling problem with
minimizing two criteria simultaneously. Liu et al. (2013) used SA for multi-criteria optimization
problem. Jolai et al. (2013) presented three multi-objective based simulated annealing algorithms to
solve a no-wait two-stage flexible flow shop scheduling problem with a number of identical machines
at each stage. Baños et al. (2013) presented a multi-objective procedure based on simulated annealing
for the capacitated vehicle routing problem with time windows. Yannibelli and Amandi (2013)
integrated the multi-objective simulated annealing algorithm into the multi-objective evolutionary
algorithm for solving the project scheduling problem. Lin and Ying (2013) presented a bi-objective
multi-start simulated-annealing algorithm for permutation flow shop scheduling problems. Sahin
(2011) proposed a SA algorithm for bi-objective facility layout problem. Cakir et al. (2011) proposed a
new approach based on a simulated annealing algorithm for multi-objective optimization of single
model stochastic assembly line balancing problem.
The SA algorithm consists of four main components (Jolai et al., 2013):
(i) Configurations;
(ii) Re-configuration technique;
(iii) Cost function;
(iv) Cooling schedule.
SA uses a stochastic approach to direct the search. It allows the search to proceed to a
neighboring state even if the move causes the value of the objective function to become worse. It
guides the original local search method in the following way. If a move to a neighbor X' in the
neighborhood N(X) decreases the objective function value, or leaves it unchanged, then the move is
always accepted. More precisely, the solution X' is accepted as the new current solution if <0, where
=C(X') – C(X) and C(X) is the value of the objective function. Moves, which increase the objective
function value, are accepted with a probability of e
-(/T)
to allow the search to escape a local optimum,
where T is a parameter named as temperature. The value of T varies from a relatively large value to a
small value close to zero. These values are controlled by a cooling schedule that specifies the initial

and incremental temperature values at each stage of the algorithm. The details of SA algorithm is
mentioned in (Tavakkoli-Moghaddam et al., 2011; Abbasi et al., 2011; Ribeiro et al., 2011; Xhafa et
al., 2011).
4. An Overview about Scatter Search
Scatter search (SS) is an evolutionary method that has been successfully applied to hard optimization
problems. The fundamental concepts and principles of the method were first proposed in the 1970s,
based on formulations dating back to the 1960s to combine decision rules and problem constraints. In
contrast to other evolutionary methods like genetic algorithms, scatter search is founded on the premise
that systematic designs and methods for creating new solutions afford significant benefits beyond those
derived from recourse to randomization. It uses strategies for search diversification and intensification
that have proved effective in a variety of optimization problems (Marti et al., 2006). Scatter search is a
population based meta-heuristic method that uses a reference set to combine its solutions and construct
others. Scatter search is an evolutionary method that has been successfully applied to hard optimization
problems. The key idea of SS in multi-objective optimization is to maintain a good search time to
balance the solutions’ concentration and diversity so that the solutions on the Pareto border has certain
dispersion (Zhang et al., 2012). Diaz et al. (2006) performed an empirical evaluation of the
effectiveness of different ways of parallelizing the SS algorithm. They identified two sets of tasks
within the SS template that can be parallelized, and to design several simple, intuitive ways of carrying
out this parallelization. Gortazar et al. (2010) adapted SS to develop a black box solver for
optimization problems with binary variables. Herrera et al. (2006) deal with a continuous version of the
scatter search, which works directly with vectors of real components. Ibaneez et al. (2012) proposed a
new skull-face overlay method based on the scatter search evolutionary algorithm. The SS framework
has been widely studied because it is very flexible and can be easily applied to many diverse
optimization problems. For instance, some steps of the SS framework can be omitted or the
procedures can be rearranged. In the recent years, the SS framework has been widely used to solve
many complex combinatorial optimization problems, and it was shown to be more effective than other
metaheuristics in many instances (Zhang et al., 2012). Marti (2006) reviewed the scatter search
implementations and developments from 1994 to 2006. He also showed the cumulative frequency,
which shows the dramatic increase in the number of publications dealing with SS. From these
applications: methodology, assignment, binary problems, clustering/selection, coloring, commercial
soft, continuous, graph drawing, graph problems, knapsack, linear ordering, mixed integer
programming, multiobjective, neural networks, p-Median, permutation problems, routing, scheduling,
and software testing. Marti (2002) proposed a SS to solve a real-life problem with multiple objectives.
Pendharkar (2013) presented an interactive multi-criteria procedure that combine scatter and random
search for coal production planning problem with fuzzy profit and fuzzy coal quality decision-maker
utilities. Silva et al. (2013) proposed an improved scatter search to deal with multiobjective
environmental/economic dispatch problems based on concepts of Pareto dominance and crowding
distance and a new scheme for the combination method.
A scatter search implementation consists of the following five methods (Marti et al., 2006):
1) A diversification generation method to generate a collection of diverse trial solutions,
using an arbitrary trial solution (or seed solution) as an input.
2) An improvement method to transform a trial solution into one or more enhanced trial
solutions. (Neither the input nor the output solutions are required to be feasible, though
the output solutions will more usually be expected to be so. If no improvement of the
input trial solution results, the ‘‘enhanced” solution is considered to be the same as the
input solution.)
3) Reference set update method to build and maintain a reference set consisting of the b
‘‘best” solutions found (where the value of b is typically small, e.g., no more than 20),

organized to provide efficient accessing by other parts of the method. Solutions gain
membership to the reference set according to their quality or diversity.
4) A subset generation method to operate on the reference set, to produce a subset of its
solutions as a basis for creating combined solutions.
5) Solution combination method to transform a given subset of solutions produced by the
Subset Generation Method into one or more combined solution vectors.
5. The Proposed Hybrid SS-SA Approach
In this section we describe the proposed approach SS-SA.
5.1. The Diversification Generation Method
This method is applied to generate diverse solutions. In the proposed hybrid approach, the initial
population is generated using the diversification generation method. The weighted method is used to
generate diverse solutions. The weighted method generates 100 solutions; each solution generated is
checked for feasibility although the SS accept infeasible solutions in the first phase but we check
feasibility at the first phase to save time in the coming phases. Diversification generation method can
construct initial trial solution set quickly. Most SS algorithms generally use heuristics to obtain initial
trial solutions and adopt methods with few runtime to improve solutions. The procedure of generating
new solutions is usually developed according to the characteristic of problem, which is the essential of
an SS algorithm. However, generating the initial trial solutions and the methods of improving solution
are important in SS (Tang et al., 2010).
5.2. The Improvement Method
The improvement method is generally used after generating new solutions. The new solutions are the
initial trial solutions or solutions generated by combination method. Hence, the improvement method
need treat with feasible and infeasible solutions. But in our proposed hybrid approach we check
feasibility and select only the feasible solution to save time and find out quality solutions. In our
proposed SS-SA, the SA approach is used to improve the previously generated solutions. And from the
implementation we found that the SA leads to better solutions.
5.3. The Reference Set Update Method
The scatter search algorithm constructs and updates the refset using this method. The reference set
update method stores the solutions in the refset according to their quality to form the subset refset1 and
according to their diversity, to form the subset refset2. In order to maintain appropriate diversity in
refset1 the SA is used. The solutions in the refset1 are sorted according to the values of the objective
functions (the minimum values). The SA steps are applied on the best solution of the refset1 to
generate other good solutions. The good generated solutions replace the bad ones. As known the SA
algorithm jumps from local minimum to global minimum.
After updating refset, if there is no difference between new refset and the previous one, the
refset rebuilding approach is used. In this approach, refset2 is cleared and filled with generated
solutions using the diversification generation method.
Reference set update method completes building the initial reference set and updating the
reference set when the new solution is created.
5.4. The Subset Generation Method
Before the combination method is used to generate the new solutions, the subset generation method
forms subsets which are used in the combination method. The proposed subset generation method

divides refset solutions into three kinds of desired subsets. Each of these subsets consists of one or two
solutions. In our implementation, we use 2-element subsets as a type of subsets.
5.5. The Combination Method
In order to generate new solutions, two methods are used; the SA method and two-point crossover. We
check feasibility before selecting the new solution. We found that the solutions generated using SA are
better than the second method.
6. Computational Analysis
To evaluate our proposed approach, test problems are used and solved using the proposed approach.
These test problems are solved in the literature so we compare the results of our proposed approach
with the results of the other approaches used these problems. The proposed hybrid approach is coded
using Matlab version R2010a and executed on Intel(R) Core(TM) i3, 2.13 GHz, Windows 7 using 2
GB of RAM.
6.1. The Test Problems
These test problems are used in most of the research in the literature such as Caponio et al. (2007); Zou
et al. (2013); Yang et al. (2013); Enayatifar (2013); Gong et al. (2013); Jiao et al. (2013); Kaveh and
Laknejadi (2011); Beausoleil (2006); Deb et al. (2002). Table 1 shows the test problems.
Table 1: Test problems
Prob. N Bounds Objective functions Solutions Type
ZDT1 30 [0, 1] Convex
ZDT2 30 [0, 1] Nonconvex
ZDT3 30 [0, 1]
Convex,
disconnected
ZDT4 10 Nonconvex

Table 1: Test problems - continued
ZDT6 10 [0, 1]
Nonconvex,
nonuniformly spread
6.2. Comparison Metrics
It is difficult to compare results of one multi-objective method to another, as there is not a unique
optimum in multi-objective optimization as in single objective optimization. So, the best solution in
multi-objective terms is decided by the decision maker (Suman and Kumar, 2006). So there are some
comparison metrics that are used to compare among developed approaches. Rahimi-Vahed et al.,
(2008) mentioned five comparison metrics. They are the number of Pareto solutions, error ratio,
generational distance, spacing metric and diversification metric. Where:
1. The number of Pareto solutions. This metric shows the number of Pareto optimal solutions
that each algorithm can find. The number of found Pareto solutions corresponding to each
algorithm is compared with the total Pareto-optimal solutions obtained by the total
enumeration algorithm.
2. Error ratio. This metric allows us to measure the non-convergence of the algorithms towards
the Pareto-optimal frontier. The error ratio is given by:
N
e
E
n
i i 
 1
Where N is the number of found Pareto-optimal solutions, and ei is zero if the solution i 
Pareto optimal frontier and 1 otherwise. The closer this metric is to 1, the less the solution has
converged toward the Pareto-optimal frontier.
3. Generational distance (GD). This metric allows measurement of the distance between the
Pareto-optimal frontier and the solution set. The definition of this metric is:
N
d
GD
n
i i 
 1
Where di is the Euclidean distance between solution i and the closest which belongs to the
Pareto-optimal frontier obtained from the total enumeration.
4. Spacing metric (S). The spacing metric gives a measure of the uniformity of the spread of
points of the solution set. The metric of spacing [233] gives an indication of how evenly the
solutions are distributed along the discovered front.
It is given by:
2/1
1
2
)(
1
1








 
n
i
idd
N
S
Where d is the mean value of all di .
5. Diversification metric (D). This metric measures the spread of the solution set and is defined as:
||)ýmax(||
1
i
N
i
ixD  
Where ||ý|| iix  is the Euclidean distance between the non-dominated solution ix and the
nondominated solution iý

6.3. Parameter Setting
We have run the proposed approach with different values for the parameters of SS and SA and we
reached to the following parameters values to get quality solutions:
 The population size is set to 100 solutions
 The maximum size of RefSet1, b1, is set to 30 and the maximum size of RefSet2, b2, is set to 30.
T0=0.95 ,  = 0.95, Tf = 0.05, N (number of iterations at each temperature) = 20
6.4. Results Comparisons
We solved all the test problems and the following tables show the performance comparison between
our proposed hybrid approach and other approaches mentioned in (Kaveh and Laknejadi, 2011) for
some of the problems stated in table 1. Table 2 shows the results for the problem ZDT1, Table 3 shows
the results for the problem ZDT3, Table 4 shows the results for the problem ZDT4, Table 5 shows the
results for the problem ZDT6. These tables show the mean value, standard deviation (s) and the time
required per run for the metrics mentioned in section 5.2. From the comparison, we see that our
proposed approach perform well for most of the problems. For some problems it gives results near to
the other approaches compared with.
Table 2: Results for problem ZDT1
NSG-II PEAS MOPSO CSS-MOPSO SS-SA
GD (mean) 0.003731 0.004932 0.096400 0.003048 0.004230
GD (s) 0.000342 0.006013 0.011428 0.000288 0.000372
S (mean) 0.503569 3.765871 0.756200 0.199631 0.216251
S (s) 0.052127 1.367000 0.145703 0.048466 0.061742
D (Mean) 1 0.901099 0.725329 1 1
D (s) 0 0.077259 0.020220 0 0
Time (min) 0.551 0.029 0.091 0.098 0.123
GD (mean) 0.005031 0.082004 0.068005 0.004781 0.00523
GD (s) 0.000162 0.107889 0.013964 0.000141 0.000197
S (mean) 0.502427 2.044142 0.794325 0.301347 0.414025
S (s) 0.047587 1.500228 0.070546 0.042321 0.050318
D (Mean) 0.929290 0.583419 0.695206 0.929012 0.825601
D (s) 0.000438 0.177700 0.048935 0.000846 0.007034
Time (min) 1.019 0.052 0.213 0.216 0.411
GD (mean) 0.003559 0.450385 0.228776 0.003462 0.00169
GD (s) 0.000589 0.063752 0.070243 0.000380 0.003215
Table 4: Results for problem ZDT4 - continued
S (mean) 0.485384 2.799198 1.081437 0.175199 0.29452
S (s) 0.052186 2.015591 0.195925 0.031783 0.04320
D (Mean) 1 13.36665 3.430176 0.999987 2.06781
D (s) 0 3.921435 2.176589 0.000071 0.50673
Time (min) 1.282 0.049 0.251 0.213 0.518

GD (mean) 0.024666 0.011739 0.019078 0.026345 0.030612
GD (s) 0.024338 0.016882 0.009291 0.014011 0.019813
S (mean) 2.151897 3.499715 3.707637 3.179233 4.017011
S (s) 2.285011 2.217825 0.849501 1.351519 2.001026
D (Mean) 1 0.933755 0.903402 1 0.986012
D (s) 0 0.126736 0.267582 0 0.193544
Time (min) 1.032 0.037 0.101 0.119 0.986
7. Conclusions
In this paper we have presented a hybrid approach based o scatter search and simulated annealing to
solve the multi-objectives optimization problems. Different test problems were used to compare the
performance of our approach with other approaches. The results show that our proposed approach is
effective and competitive with the other developed approaches in the literature.
Our recommendations are to develop studies for determining the best values for the SS and SA
parameters, dealing with dynamic multiobjective optimization problems and stochastic multiobjective
optimization problems.
References
[1] Abbasi, B., Niaki, S., Abdi Khalife, M., Faize, Y., 2011. A hybrid variable neighborhood
search and simulated annealing algorithm to estimate the three parameters of the Weibull
distribution, Expert Systems with Applications 38, pp. 700–708.
[2] Bajestani, M. A., Rabbani, M., Rahimi-Vahed, A.R., Khoshkhou, G. B., 2009. A multi-
objective scatter search for a dynamic cell formation problem, Computers & Operations
Research, 36, pp. 777 – 794.
[3] Bandyopadhyay, S., Saha, S., Maulik, U., Deb, K., 2008. A Simulated Annealing-Based
Multiobjective Optimization Algorithm: AMOSA, IEEE Transactions on evolutionary
computation, 12 (3), pp. 269-283
[4] Baños, R., Ortega, J., Gil, C., Fernández, A., Toro, F., 2013. A Simulated Annealing-based
parallel multi-objective approach to vehicle routing problems with time windows, Expert
Systems with Applications, 40 (5), pp. 1696-1707.
[5] Beausoleil, R. P., 2006. "MOSS" multiobjective scatter search applied to non-linear multiple
criteria optimization, European Journal of Operational Research, 169, pp. 426–449.
[6] Cakir, B., Altiparmak, F., Dengiz, B., 2011. Multi-objective optimization of a stochastic
assembly line balancing: A hybrid simulated annealing algorithm, Computers & Industrial
Engineering, 60, pp. 376–384.
[7] Caponio, A., Cascella, G.L., Neri, F., Salvatore, N., Sumner, M., 2007. A fast adaptive
memetic algorithm for on-line and off-line control design of pmsm drives. IEEE Transactions
on System Man and Cybernetics-part B, special issue on Memetic Algorithms 37(1), pp. 28–41.
[8] Deb K., Pratap, A., Agarwal, S., T. Meyarivan, 2002. A Fast and Elitist Multiobjective Genetic
Algorithm: NSGA-II, IEEE Transactions on Evolutionary Computation, 6 (2), pp. 182-197.
[9] Diaz, B., Carbajal, S., Lozano, S., 2006. An empirical investigation on parallelization strategies
for Scatter Search, European Journal of Operational Research, 169, pp. 490–507.
[10] Ekbal, A., Saha, S., 2011. A multiobjective simulated annealing approach for classifier
ensemble: Named entity recognition in Indian languages as case studies, Expert Systems with
Applications, 38, pp. 14760–14772.
[11] Enayatifar,R.,Yousefi,M.,Abdullah,A.,Darus,A.N.,2013.MOICA:Anovelmulti-objectiveapproachbasedon
imperialistcompetitivealgorithm,AppliedMathematicsandComputation,219,pp.8829–8841.

[12] Gong, D., Sun, J., Ji, X., 2013. Evolutionary algorithms with preference polyhedron for interval
multi-objective optimization problems, Information Sciences, 233, pp. 141–161.
[13] Gortazar, F., Duarte, A., Laguna, M., Marti, R., 2010. Black box scatter search for general classes of
binary optimization problems, Computers & Operations Research, 37, pp. 1977–1986.
[14] Herrera, F., Lozano, M., Molina, D., 2006. Continuous scatter search: An analysis of the
integration of some combination methods and improvement strategies, European Journal of
Operational Research, 169, pp. 450–476.
[15] Ho, S. L., Yang, S., Wong, H. C., and Ni, G., 2003. A Simulated Annealing Algorithm for
Multiobjective Optimizations of Electromagnetic Devices, IEEE Transactions on Magnetics, 39
(3), pp. 1285-1288.
[16] Ibaneez, O., Cordon, O., Damas, S., Santamaria, J., 2012. An advanced scatter search design
for skull-face overlay in craniofacial superimposition, Expert Systems with Applications, 39,
pp. 1459–1473.
[17] Jarosław, P., Czesław, S., Dominik, Z., 2013. Optimizing bicriteria flow shop scheduling
problem by simulated annealing algorithm, Procedia Computer Science, 18, pp. 936 – 945.
[18] Jiao, L.C., Wang, H., Shang, R.H., Liu, F., 2013. A co-evolutionary multi-objective
optimization algorithm based on direction vectors, Information Sciences, 228, pp. 90–112.
[19] Jolai, F., Asefi, H., Rabiee, M., Ramezani, P., 2013. Bi-objective simulated annealing
approaches for no-wait two-stage flexible flow shop scheduling problem, Scientia Iranica E, 20
(3), pp. 861–872.
[20] Kaveh A., Laknejadi K., 2011. A novel hybrid charge system search and particle swarm optimization
method for multi-objective optimization, Expert Systems with Applications, 38, pp. 15475–15488.
[21] Konak A., Coit, D. W., Smith, A. E., 2006. Multi-objective optimization using genetic
algorithms: A tutorial, Reliability Engineering and System Safety, 91, pp. 992–1007.
[22] Li, X., Du, G., 2013. BSTBGA: A hybrid genetic algorithm for constrained multi-objective
optimization problems, Computers & Operations Research, 40, pp. 282–302.
[23] Lin, S., Ying, K-C., 2013. Minimizing makespan and total flow time in permutation flow shops
by a bi-objective multi-start simulated-annealing algorithm, Computers & Operations Research,
40 (6), pp. 1625-1647.
[24] Liu, L., Mu, H., Yang, J., Li, X., Wu, F., 2013. A simulated annealing for multi-criteria
optimization problem: DBMOSA, Swarm and Evolutionary Computation, 14, pp.48 -65.
[25] Marler R.T. and Arora J.S., 2004. Survey of multi-objective optimization methods for
engineering, Struct Multidisc Optim, 26, 369–395.
[26] Marti, R., 2006. Scatter Search—Wellsprings and Challenges, European Journal of Operational
Research, 169, pp. 351–358.
[27] Marti R, Laguna M, Glover F., 2006. Principles of scatter search. European Journal of
Operational Research, 169, pp. 359–372.
[28] Miettinen, K., M., 1999. Nonlinear Multiobjective Optimization, Kluwer Academic Publishers, Boston.
[29] Moghaddam, R., Makui, A., Mazloomi, Z., 2010. A new integrated mathematical model for a
bi-objective multi-depot location-routing problem solved by a multi-objective scatter search
algorithm, Journal of Manufacturing Systems, 29, pp. 111–119.
[30] Molina, J., Laguna, M., Mart, R., and Caballero, R., 2007. SSPMO: A Scatter Tabu Search Procedure for
Non-LinearMultiobjectiveOptimization,INFORMS Journal on Computing, 19 (1), pp. 91–100.
[31] Mousa, A.A., El-Shorbagy, M.A., Abd-El-Wahed, W.F., 2012. Local search based hybrid
particle swarm optimization algorithm for multiobjective optimization, Swarm and
Evolutionary Computation, 3 pp.1–14.
[32] Pendharkar, P., 2013. Scatter search based interactive multi-criteria optimization of fuzzy
objectives for coal production planning, Engineering Applications of Artificial Intelligence, 26,
pp. 1503-1511.

[33] Rahimi-Vahed A.R., Javadi B., Rabbani M. and Tavakkoli-Moghaddam R., 2008. A multi-
objective scatter search for a bi-criteria no-wait flow shop scheduling problem, Engineering
Optimization, 40 (4), pp. 331–346.
[34] Rahimi-Vahed, A., Rabbani, M., R. Moghaddam, T., Torabi, S.A., Jolai, F., 2007. A multi-
objective scatter search for a mixed-model assembly line sequencing problem, Advanced
Engineering Informatics, 21, pp. 85–99
[35] Ribeiro, G. M., Mauri, G. R., Lorena, L. A., 2011. A simple and robust Simulated Annealing
algorithm for scheduling workover rigs on onshore oil fields, Computers & Industrial
Engineering, 60, pp. 519–526
[36] Sahin, R., 2011. A simulated annealing algorithm for solving the bi-objective facility layout
problem, Expert Systems with Applications, 38, pp. 4460–4465
[37] Silva, C., Climaco, J., Figueira, J., 2006. A scatter search method for bi-criteria {0,1}-knapsack
problems, European Journal of Operational Research, 169, pp. 373–391
[38] Silva, M., Klein, C. E., Mariani, V. C., Coelho, L., 2013. Multiobjective scatter search
approach with new combination scheme applied to solve environmental/economic dispatch
problem, Energy, 53, pp. 14-21
[39] Suman, B., and Kumar, P., 2006. A survey of simulated annealing as a tool for single and
multiobjective optimization, Journal of the Operational Research Society, 57, pp. 1143–1160
[40] Talaslioglu, T., 2010. MultiObjective Design Optimization of Grillage Systems by Scatter
Search Methodology, International Journal of Civil and Structural Engineering, 1 (3), pp. 477-
496
[41] Tang, J., Zhang, J., Pan, Z., 2010. A scatter search algorithm for solving vehicle routing
problem with loading cost, Expert Systems with Applications, 37, pp. 4073–4083
[42] Tavakkoli-Moghaddam, R., Gazanfari, M., Alinaghian, M., Salamatbakhsh, A., Norouzi, N.,
2011. A new mathematical model for a competitive vehicle routing problem with time windows
solved by simulated annealing, Journal of Manufacturing Systems, 30, pp. 83– 92
[43] Xhafa, F., Barolli, A., Sanchez, C., Barolli, L., 2011. A simulated annealing algorithm for
router nodes placement problem in Wireless Mesh Networks, Simulation Modelling Practice
and Theory, 19, pp. 2276–2284
[44] Yang, X., Karamanoglu, M., He, X., 2013. Multi-objective Flower Algorithm for Optimization,
Procedia Computer Science, 18, pp. 861 – 868
[45] Yannibelli, V., Amandi, A., 2013. Hybridizing a multi-objective simulated annealing algorithm
with a multi-objective evolutionary algorithm to solve a multi-objective project scheduling
problem, Expert Systems with Applications, 40 (7), pp. 2421-2434
[46] Zhang, T., Chaovalitwongse, W., Zhang, Y., 2012. Scatter search for the stochastic travel-time
vehicle routing problem with simultaneous pick-ups and deliveries, Computers & Operations
Research 39, pp. 2277–2290
[47] Zou, F., Wang, L., Hei, X., Chen, D., Wang, B., 2013. Multi-objective optimization using
teaching-learning-based optimization algorithm. Engineering Applications of Artificial
Intelligence, 26, pp. 1291–1300

EJSR(5)

Recommended

Recommended

More Related Content

What's hot

What's hot (16)

Viewers also liked

Viewers also liked (9)

Similar to EJSR(5)

Similar to EJSR(5) (20)

More from AMIT KUMAR

More from AMIT KUMAR (9)

EJSR(5)