
Research Article


Ant Colony Optimization: Approach for an Obstacle 

Munir A. Ghanem
and
Rabei A. Ahmed



ABSTRACT

This study presents a robust optimization algorithm based on updating Ant Colony Optimization (ACO) through hybridization with Artificial Bee Colony (ABC) method and information exchange concept for the purpose of covering ACO limitations in case of an obstacle was on the ant’s path to food source. The global optimal solution found by the proposed hybrid ACO and ABC (ACOBC) algorithm is considered to be as novel technique to find the shortest path when the vision to food source location is not clear because of an obstacle. Both of the ACO and ABC methods share the globally best solutions through the information exchange process between the ants and bees. Based on the results, the exchange process significantly increases exploration and exploitation of the hybrid method. Besides, a focused elitism scheme is introduced to enhance the performance of the developed algorithm. The validity of the ACOBC method is verified using several continuous test problems and a typical discrete problem, called Traveling Salesman Problem (TSP). The proposed method is found to be a competitive optimization tool for solving continuous and discrete problems. Obstacle model study is very important due to its significance in solving many complex networking problems connected to real human life situations where real data are not available.





Received: October 28, 2015;
Accepted: December 28, 2015;
Published: January 15, 2016


INTRODUCTION Optimization is the choice of a vector for an objective function in a given domain to make an optimal solution. In the last two decades, several metaheuristic techniques have been developed to solve difficult optimization problems. Some of these problems are reliability (Zou et al., 2010, 2011a), feature selection (Li and Yin, 2013a), knapsack (Zou et al., 2011b), face detection (Owusu et al., 2014), scheduling (Li and Yin, 2013b), pattern recognition (Kumar and Rani, 2013), economic load dispatch (Zhisheng, 2013), classification (Diao et al., 2012) and image segmentation (Zhang et al., 2011). The potential of metaheuristic optimization approaches for addressing various maximization/minimization problems is wellunderstood. This is evident from the sizeable number of recently proposed modern stochastic optimization methods (Yang et al., 2012; Gandomi et al., 2013d). Some of the major metaheuristic optimization methods that have been applied to solve challenging optimization problems are: Differential Evolution (DE) (Storn and Price, 1997; Gandomi et al., 2012; Li and Yin, 2012), Bat Algorithm (BA) (Yang and Gandomi, 2012; Gandomi et al., 2013c; Yang, 2010; Mirjalili et al., 2014c), genetic algorithms (GAs) (Goldberg, 1989; Manurung et al., 2012), Evolutionary Strategy (ES) (Beyer and Schwefel, 2002), Artificial Bee Colony (ABC) (Karaboga and Basturk, 2007), Genetic Programming (GP) (Gandomi and Alavi, 2011), fruit fly optimization algorithm (Pan, 2012), Animal Migration Optimization (AMO) (Li et al., 2014), Artificial Plant Optimization Algorithm (APOA) (Cai et al., 2012), artificial physics optimization (Xie et al., 2012), ProbabilityBased Incremental Learning (PBIL) (Baluja, 1994), Grey Wolf Optimizer (GSO) (Mirjalili et al., 2014a), BiogeographyBased Optimization (BBO) (Simon, 2008; Mirjalili et al., 2014b), Harmony Search (HS) (Geem et al., 2001; Yadav et al., 2012), Flower Pollination Algorithm (FOA) (Yang et al., 2014), Particle Swarm Optimization (PSO) (Mirjalili and Lewis, 2013; Kennedy and Eberhart, 1995; Talatahari et al., 2013; Mirjalili et al., 2012), Charged System Search (CSS) (Kaveh and Talatahari, 2010), Ant Colony Optimization (ACO) (Dorigo and Stutzle, 2004) and Cuckoo Search (CS) (Gandomi et al., 2013a, b; Yang and Deb, 2009). Ant Colony Optimization (ACO) is a metaheuristic in which a colony of artificial ants cooperates in finding good solutions to difficult discrete optimization problems. The ACO obstacle approach problem is a good model for combinational optimization, where many of such problems are considered to be as NPhard, though it is very important to find very quick and highquality solution (Dorigo and Stutzle, 2004). On the other side, the ABC algorithm, motivated by the swarm behaviours of bee colonies has a quite simple yet effective structure for solving optimization problems (Karaboga and Basturk, 2007). Hence, it has attracted the attention of many researchers. It is known that the metaheuristic methods require various exploration and exploitation schemes for solving problems with increasing dimensions in the search space. Although, the ACO generally explores the search space well and appears to be fully capable of locating the global optimal value, its exploration ability has exhibited relatively poor performance at later run phase; especially in the obstacle case where the food source location on the other side of the obstacle is obscure and in the case of arising new shorter path but the ants won’t change its path (Dorigo and Stutzle, 2004). On the other hand, the ABC method has strong exploration ability with its poor exploitation (Zhu and Kwong, 2010). Therefore, single ACO or ABC method seems not to be efficient for the exploration and exploitation of the search space. To cope with this issue, this study presents a hybridization of the ABC and ACO methods for solving continuous numerical global optimization as well as discrete problems that optimally solves obstacle model.
MATERIALS AND METHODS ACO method: Examples of metaheuristics include simulated annealing (Cerny, 1985; Kirkpatrick et al., 1983), tabu search (Glover, 1989, 1990; Glover and Laguna, 1997), iterated local search (Lourenco et al., 2002), evolutionary computation (Fogel et al., 1966; Holland, 1975; Rechenberg, 1973; Schwefel, 1981; Glover, 1989). In formally, an ACO algorithm can be imagined as the interplay of three procedures. Construct ants solutions, update pheromones and Daemon actions, where the work of every procedure is clear from its name without going into more details. Many ACO algorithms have been suggested in this field, we will proceed with the most successful algorithm; "Ant Colony System (ACS)" (Dorigo et al., 2006). The ACS mechanism adds a local pheromone (update) at the end of each constructed solution step. Local pheromone update is performed by all ants to diverse solution in every iteration through decreasing pheromone and ultimately to give subsequent ants a chance to search for different solutions. The formula of local pheromone updated in Eq. 1: where, φ denotes pheromone evaporation and μ_{0} is the initial value of pheromone. In addition to local pheromone update, ACS performs pheromone update at the end of each iteration by only one ant which can be either the iterationbest or the bestsofar. Update formula in Eq. 2:
where, Δμ_{kl} = 1/L_{best} and L_{best} is the tour length of best ant that can be found either in the current "Iterationbest" or "Bestsofar" or combination of both. ACO obstacle approach problem is a good model for combinational optimization, where many of such problems are considered to be as NPhard, though it is very important to find very quick and highquality solution.
ABC method: Artificial Bee Colony is one of the seminal metaheuristic methods among various intelligent optimization techniques. After the appearance of swarm intelligence of bee colony, the forage selection is modeled. Based on this model, the definition of three main concepts can be defined as follows (Zhang and Wu, 2012):
• 
Food resource: In the simplest form, the value of a food source is described with only one quantity. Figure 1 represents two food resources and two nonfood resources, respectively. Furthermore, S, O, R, UF and EF denote scouts, onlookers, recruits, unemployed foragers and denote employed foragers, respectively 
• 
Unemployed foragers: The unemployed forages have two sorts. One is Scouts (S). A scout bee is type of bee that begins implementing search autonomously without any a priori knowledge 
The other one is Onlookers (O). They only stay in the nest in order to search for a food source with the help of the employed foragers.
• 
Employed foragers: All of them are related to a food source that they are exploiting now. This information is shared with some probability. Three feasible choices associated with the quantity of nectar are provided for the foraging bee 
• 
One is Unemployed Forager (UF): When the nectar is less than a fixed threshold, the foraging bee gives it up and turns to an unemployed bee 
• 
Employed Forager 1 (EF1): If not, it may dance and recruit mates. The last one is Employed Forager 2 (EF2). It may forage around the food source all the time (Karaboga and Basturk, 2008) 
The artificial bee colony includes three types of bees: (1) Employed bees, (2) Onlookers and (3) Scouts. In the artificial bee colony, a food source corresponds to an employed bee. That is to say, the employed bees and the food sources have the same number. The detailed main steps of the search conducted by the artificial bees can be described as follows (Sabat et al., 2010):
Step 1: 
Initialize the population x_{ij} 
Step 2: 
Repeat 
Step 3: 
Generate new solutions v_{ij} around x_{ij} for the employed bees as in Eq. 3: 
v_{ij} = x_{ij}+Φ (x_{ij}x_{kj}) 
(3) 
Here, k is a solution around i, Φ is a random number [1, 1].
Step 4: 
The greedy selection is used between x_{i} and v_{i} 
Step 5: 
Calculate the probability P_{i} for x_{i} according to their fitness as in Eq. 4: 
SN is the number of food sources and f_{i} is its fitness.
Step 6: 
Normalize P_{i} into [0, 1] 
Step 7: 
Generate the new solutions v_{i} for the onlookers from x_{i}, selected depending on P_{i} 
Step 8: 
The greedy selection is used for the onlookers between x_{i} and v_{i} 
Step 9: 
Check if a solution is abandoned. If it is, replace it with a novel one x_{i} for the scout as in Eq. 5: 
x_{ij} = min_{j}+Φ_{ij} (max_{j}min_{j}) 
(5) 
Here, Φ_{ij} is a random number in [0, 1].
Step 10: 
Save the best solution obtained up to now 
Step 11: 
Go to Step 2 until termination criteria is satisfied 
ACOBC method: Based on the aboveanalyses of the ACO, when the ants gets trapped in the local values, it cannot escape from local minimum by itself, also the ants cannot know by itself the shorter path to food source located on the other side of the obstacle. Further, ABC algorithm does not directly utilize the global optimal individual. To overcome these limitations, a hybrid metaheuristic method based on information exchange is presented. The hybridization process is similar to that proposed by Kiran and Gunduz (2013). Information exchange or crossover operation is one of the most famous evolution operators. Here, it is used for yielding a new solution, called The Best. The Best is considered to be Abest for the ACO and food source of onlooker bees for the ABC. To get TheBest, the Abest of the ACO and the optimal individual of the ABC are computed by Eq. 4. Probabilities used to select the two solutions are given by Eq. 6 and 7: where, P_{best} is the probability to choose the optimal individual of the ABC, fit_{best} and fit_{Abest} are the Abest of the ACO and the optimal individual of the ABC achieved according to Eq. 4: where, P_{Abest} is the Abest of the ACO. When generating the best solution, random numbers in the range of [0, 1] are utilized for the dimensions of the standard test function. If it is not above P_{best}, the value for this dimension is selected from the optimal individual of ABC. Otherwise, this value is selected from Abest of ACO. This selection process can be formulated as Eq. 8:
where, The Best_{i} is the ith dimension of The Best. The Best_{i} is ith dimension of The Best solution found by ABC, A best_{i }is the ith dimension of Abest of the ACO. The r is a random number in the range of [0, 1]. Based on the information exchange described above, the connection between the ants and bees in the ACOBC method can be stated as follows:
• 
The global part of the ACOBC method is "The Best". Through The Best, not only ACO has enhanced the ability of escaping from local minima but also the exploitation of ABC is significantly enhanced by the direct utilization of the global best solution. 
• 
Abest of the ACO is updated in terms of The Best accordingly and the same is passed on to onlooker bees of ABC as neighbor. 
• 
Besides, a concentrated elitism strategy is introduced into ACOBC to forbid the optimal solutions from being ruined by the method. This is done to guarantee that the whole population is capable of proceeding with a better status than before. By introducing this concentrated elitism strategy into the algorithm, the ACOBC has been further developed. 
RESULTS AND DISCUSSION Bench mark evaluation: In order to validate ACOBC, it has been applied to optimize a series of benchmark functions from previous studies presented in Table 1 (Yang et al., 2013; Wang et al., 2014b). In order to conduct a fair comparison, all the simulations were implemented on the same environments (Wang et al., 2013; Guo et al., 2014).
Here, the performance of ACOBC was compared with nine natureinspired methods viz. ABC (Karaboga and Basturk, 2007), ACO (Dorigo and Stutzle, 2004), DE (Storn and Price, 1997), ES (Beyer, 2001), GA (Goldberg, 1989), HS (Geem et al., 2001), KH (Mousavi et al., 2013), PBIL (Baluja, 1994) and PSO (Kennedy and Eberhart, 1995). Furthermore, Dumitrescu and Stutzle (2004) performance was the overall best performance, with respect to the number of best known solutions found was obtained by ACS. Therefore, ACS is selected as the representative of ACO algorithm. For ACO and ACOBC, the same parameters are suggested as well as for the other methods parameters are set suggested in the study of Wang et al. (2014ac). It is wellknown that most of the metaheuristic methods are based on certain type of stochastic distribution. To obtain typical performances, threehundred trials are implemented for each method on each function (Table 2 and 3). Different standards are considered to normalize values in the tables. Therefore, values are not comparative between each other (Wang et al., 2014c). The dimension of the benchmark is set thirty. From Table 2, it can be seen that ACOBC has the best performance on nineteen of the twentyfive test problems. Furthermore, the performance of DE is only worse than ACOBC. For best solutions shown in Table 3, ACOBC provides the best results for twentyone of the twentyfive test problems. TSP problem: Traveling Salesman Problem (TSP) is a typical NPcomplete problem. It is difficult to solve this problem using traditional methods. The TSP is not merely the traveling salesman problem. Many other NP problems can be attributed to TSP such as postman problem, nut production scheduling problem and product assembly line. Therefore, the study of TSP is of great importance. The distance between each of n cities or their coordinates are provided. A traveling salesman starts to visit each city once and only once from certain city and finally returns to the starting city. The task involves how to arrange this traveling in order to make the shortest route. In short, TSP is to find a shortest trajectory among n cities or search for a city permutation π(X) = {V_{1}, V_{2}, …, V_{n}} in a natural subset X = {1, 2, … , i , … , n}.
Table 1:  Benchmark functions  
Table 2: 
ACOBC mean function values compared with other functions 

ABC: Artificial bee colony, ACO: Ant colony optimization, DE: Differential evolution, ES: Evolutionary strategy, GA: Genetic algorithm, HS: Harmony search, PBIL: Probability based incremental learning, PSO: Particle swarm optimization 
Table 3: 
ACOBC function values compared with other functions 

ABC: Artificial bee colony, ACO: Ant colony optimization, DE: Differential evolution, ES: Evolutionary strategy, GA: Genetic algorithm, HS: Harmony search, PBIL: Probability based incremental learning, PSO: Particle swarm optimization 
Here, i represents city number which must be integer number and its range varies from 1 to n. In other words, we minimize the total distance as represented in Eq. 9:
where, d(V_{i}, V_{i+1}) is the distance between city V_{i} and city V_{i+1}.
Table 4:  Solution for CTSP problem   ABC: Artificial bee colony, ACO: Ant colony optimization, DE: Differential evolution, ES: Evolutionary strategy, GA: Genetic algorithm, HS: Harmony search, PBIL: Probability based incremental learning, PSO: Particle swarm optimization 
In this study, a particular kind of TSP, called Chinese TSP (CTSP) problem is considered. In CTSP, there are 31 main cities in China and the coordinates of each was given. In order to prove the ability of solving discrete problem, ACOBC is applied to solve the CTSP problem. In fact, the proposed method is used to find the shortest path among these cities. Firstly, the initial paths are randomly generated as follows: 31→8→23→12→11→24→27→6→16→10→30→25→26→17→4→15→14→7→5→3→13→9→29→21→28→22→1→19→18→20→2→31 Its initial total distance is 41751 km. As it is seen, ACOBC is efficiently capable of searching for much shorter path. The final path is shown as follows: 12→13→7→10→9→8→2→4→5→16→6→11→23→24→20→19→17→18→3→22→21→25→?26→28→27→30→29→31→1→15→14→12 The calculated distance is 16087 km. This is very close to the known shortest distance 15378 km. The results are obtained with population size and maximum generation number equal to 50 and 100, respectively. Subsequently, ACOBC is compared with other methods discussed before for the CTSP problem. In order to get a fair comparison, all the methods are implemented in the same limited conditions, i.e., population size = 50 and generation number = 50. In order to remove the stochastic influence, 800 trials are conducted so as to get more accurate statistical results in Table 4. From Table 4, it can be seen that for best and mean distance, ACOBC can find the shorter path as compared to the other methods under the same conditions. Considering the worst distance, all the methods can find the similar paths except HS and PSO. In addition, ACOBC has the smallest Standard Deviation (SD) value. That is to say, ACOBC can find the paths within smaller range. Especially, ACOBC is superior to ACO that is known as one of the most efficient algorithm for the TSP problem. Based on the results, it can be concluded that ACOBC is wellsuited for solving the TSP problem. It is useful to discuss study results in comparison with other researches results that search in the same field of optimization. To summarize the comparison, ACOBC algorithm was superior due to the following reasons:
• 
No previous research discusses the main issue of this research, which is the ‘obstacle case’ 
• 
Hybridizing ACO and ABC algorithms utilizes the optimum of both algorithms to reach best of the best and it also covers limitations in both algorithms 
• 
The proposed algorithm ACOBC proves through calculations that it is superior to solve both, discrete and continuous data problems 
• 
Founding of this research is compatible and supported with all previous researches as this was proved through previous discussion 
• 
Founding of this research was compared with twenty one continuous data test problems and nine discrete data problems and found to be the best 
CONCLUSION In this study, a hybridization of the ACO and ABC methods, namely ACOBC, is proposed for the continuous and discrete optimization. Its main objective is to overcome ACO obstacle case limitations. The ACOBC integrates the capabilities of the ACO and the ABC to reach the optimal local and global solution. Moreover, a focused elitism scheme is applied to the method to further enhance its performance. This research subject is very important as it touches many of human technical key problems, such as: network routing problem, travelling salesman problem and vehicle routing problem. The results clearly demonstrate the superiority of ACOBC over ACO, ABC and other metaheuristic algorithms. However, there are quite a few issues that merit further investigation such as analyzing the parameters used in the ACOBC method. The future study can focus on solving a more ubiquitous set of different continuous optimization and discrete problems. Finally, the study of CPU time used by the metaheuristic approaches needs attention to make the proposed method more feasible for solving the practical engineering problems. ACKNOWLEDGMENTS This project was funded by deanship of Scientific Research, Northern Border University for their financial support under grant no. (7063435). The authors, therefore, acknowledge with thanks DSR technical and financial support.

REFERENCES 
1: Beyer, H.G., 2001. The Theory of Evolution Strategies. Springer, New York, USA., ISBN13: 9783662043783, Pages: 381.
2: Beyer, H. and H. Schwefel, 2002. Natural Computing. Kluwer Academic Publishers, Dordrecht Netherlands.
3: Cai, X., S. Fan and Y. Tan, 2012. Light responsive curve selection for photosynthesis operator of APOA. Int. J. BioInspired Comput., 4: 373379. CrossRef  Direct Link 
4: Diao, L., C. Yang and H. Wang, 2012. Training SVM email classifiers using very large imbalanced dataset. J. Exp. Theoret. Artif. Intell., 24: 193210. CrossRef  Direct Link 
5: Dorigo, M. and T. Stutzle, 2004. Ant Colony Optimization. MIT Press, Cambridge, MA., USA., ISBN13: 9780262042192, Pages: 305.
6: Gandomi, A.H. and A.H. Alavi, 2011. Multistage genetic programming: A new strategy to nonlinear system modeling. Inform. Sci., 181: 52275239. CrossRef  Direct Link 
7: Dorigo, M., M. Birattari and T. Stutzle, 2006. Ant colony optimization: Artificial ants as a computational intelligence technique. IRIDIATechnical Report Series, Technical Report No. TR/IRIDIA/2006023, Universite Libre De Bruxelles, Bruxelles, Belgium, September 2006.
8: Gandomi, A.H., S. Talatahari, X.S. Yang and S. Deb, 2013. Design optimization of truss structures using cuckoo search algorithm. Struct. Des. Tall Special Build., 22: 13301349. CrossRef  Direct Link 
9: Gandomi, A.H., X.S. Yang and A.H. Alavi, 2013. Cuckoo search algorithm: A metaheuristic approach to solve structural optimization problems. Eng. Comput., 29: 1735. CrossRef  Direct Link 
10: Gandomi, A.H., X.S. Yang, A.H. Alavi and S. Talatahari, 2013. Bat algorithm for constrained optimization tasks. Neural Comput. Applic., 22: 12391255. CrossRef  Direct Link 
11: Gandomi, A.H., X.S. Yang, S. Talatahari and S. Deb, 2012. Coupled eagle strategy and differential evolution for unconstrained and constrained global optimization. Comput. Math. Applic., 63: 191200. CrossRef  Direct Link 
12: Gandomi, A.H., X.S. Yang, S. Talatahari and A.H. Alavi, 2013. Metaheuristic Applications in Structures and Infrastructures. Elsevier, Waltham, MA., USA., ISBN13: 9780123983794, Pages: 568.
13: Geem, Z.W., J.H. Kim and G.V. Loganathan, 2001. A new heuristic optimization algorithm: Harmony search. Simulation, 76: 6068. CrossRef  Direct Link 
14: Goldberg, D.E., 1989. Genetic Algorithms in Search, Optimization and Machine Learning. 1st Edn., AddisonWesley Professional, Boston, MA., USA., ISBN13: 9780201157673, Pages: 412.
15: Karaboga, D. and B. Basturk, 2007. A powerful and efficient algorithm for numerical function optimization: Artificial Bee Colony (ABC) algorithm. J. Global Optim., 39: 459471. CrossRef  Direct Link 
16: Karaboga, D. and B. Basturk, 2008. On the performance of Artificial Bee Colony (ABC) algorithm. Applied Soft Comput., 8: 687697. CrossRef  Direct Link 
17: Kaveh, A. and S. Talatahari, 2010. A novel heuristic optimization method: charged system search. Acta Mech., 213: 267289. CrossRef  Direct Link 
18: Kennedy, J. and R. Eberhart, 1995. Particle swarm optimization. Proceedings of the International Conference on Neural Networks, Volume 4, November 27December 1, 1995, Perth, WA., USA., pp: 19421948.
19: Kumar, S. and A. Rani, 2013. DFLDA tree: A nonlinear multilevel classifier for pattern recognition. J. Exp. Theoret. Artif. Intell., 25: 177188. CrossRef  Direct Link 
20: Li, X. and M. Yin, 2012. Application of differential evolution algorithm on selfpotential data. PLoS ONE, Vol. 7. 10.1371/journal.pone.0051199
21: Li, X. and M. Yin, 2013. Multiobjective binary biogeography based optimization for feature selection using gene expression data. IEEE Trans. NanoBiosci., 12: 343353. CrossRef  Direct Link 
22: Li, X. and M. Yin, 2013. An oppositionbased differential evolution algorithm for permutation flow shop scheduling based on diversity measure. Adv. Eng. Software, 55: 1031. CrossRef  Direct Link 
23: Li, X., J. Zhang and M. Yin, 2014. Animal migration optimization: An optimization algorithm inspired by animal migration behavior. Neural Comput. Applic., 24: 18671877. CrossRef  Direct Link 
24: Manurung, R., G. Ritchie and H. Thompson, 2012. Using genetic algorithms to create meaningful poetic text. J. Exp. Theoret. Artif. Intell., 24: 4364. CrossRef  Direct Link 
25: Mirjalili, S. and A. Lewis, 2013. Sshaped versus Vshaped transfer functions for binary particle swarm optimization. Swarm Evolut. Comput., 9: 114. CrossRef  Direct Link 
26: Mirjalili, S., S.M. Mirjalili and A. Lewis, 2014. Grey wolf optimizer. Adv. Eng. Software, 69: 4661. CrossRef  Direct Link 
27: Mirjalili, S., S.M. Mirjalili and A. Lewis, 2014. Let a biogeographybased optimizer train your multilayer perceptron. Inform. Sci., 269: 188209. CrossRef  Direct Link 
28: Mirjalili, S., S.M. Mirjalili and X.S. Yang, 2014. Binary bat algorithm. Neural Comput. Applic., 25: 663681. CrossRef  Direct Link 
29: Mirjalili, S., S.Z.M. Hashim and H.M. Sardroudi, 2012. Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm. Applied Math. Comput., 218: 1112511137. CrossRef  Direct Link 
30: Owusu, E., Y.Z. Zhan and Q.R. Mao, 2014. An SVMAdaBoostbased face detection system. J. Exp. Theoret. Artif. Intell., 26: 477491. CrossRef  Direct Link 
31: Pan, W.T., 2012. A new fruit fly optimization algorithm: Taking the financial distress model as an example. KnowledgeBased Syst., 26: 6974. CrossRef  Direct Link 
32: Sabat, S.L., S.K. Udgata and A. Abraham, 2010. Artificial bee colony algorithm for small signal model parameter extraction of MESFET. Eng. Applic. Artif. Intell., 23: 689694. CrossRef  Direct Link 
33: Kiran, M.S. and M. Gunduz, 2013. A recombinationbased hybridization of particle swarm optimization and artificial bee colony algorithm for continuous optimization problems. Applied Soft Comput., 13: 21882203. CrossRef  Direct Link 
34: Baluja, S., 1994. Populationbased incremental learning: A method for integrating genetic search based function optimization and competitive learning. Technical Report CMUCS94163, Computer Science Department, Carnegie Mellon University, Pittsburgh, PA., USA., June 2, 1994.
35: Simon, D., 2008. Biogeographybased optimization. IEEE Trans. Evol. Comput., 12: 702713. CrossRef  Direct Link 
36: Storn, R. and K. Price, 1997. Differential evolutionA simple and efficient heuristic for global optimization over continuous spaces. J. Global Optim., 11: 341359. CrossRef  Direct Link 
37: Talatahari, S., M. Kheirollahi, C. Farahmandpour and A.H. Gandomi, 2013. A multistage particle swarm for optimum design of truss structures. Neural Comput. Applic., 23: 12971309. CrossRef  Direct Link 
38: Wang, G.G., A.H. Gandomi and A.H. Alavi, 2014. An effective krill herd algorithm with migration operator in biogeographybased optimization. Applied Math. Mod., 38: 24542462. CrossRef  Direct Link 
39: Wang, G.G., L. Guo, H. Duan and H. Wang, 2014. A new improved firefly algorithm for global numerical optimization. J. Comput. Theoret. Nanosci., 11: 477485. CrossRef  Direct Link 
40: Wang, G., L. Guo, A.H. Gandomi, L. Cao, A.H. Alavi, H. Duan and J. Li, 2013. Levyflight krill herd algorithm. Math. Problems Eng. 10.1155/2013/682073
41: Xie, L., J. Zeng and R.A. Formato, 2012. Selection strategies for gravitational constant G in artificial physics optimisation based on analysis of convergence properties. Int. J. BioInspired Comput., 4: 380391. CrossRef  Direct Link 
42: Yadav, P., R. Kumar, S.K. Panda and C.S. Chang, 2012. An intelligent tuned harmony search algorithm for optimisation. Inform. Sci., 196: 4772. CrossRef  Direct Link 
43: Yang, X.S., Z. Cui, R. Xiao, A.H. Gandomi and M. Karamanoglu, 2013. Swarm Intelligence and BioInspired Computation: Theory and Applications. Newnes, Waltham, MA., ISBN13: 9780124051775, Pages: 450.
44: Yang, X.S., M. Karamanoglu and X. He, 2014. Flower pollination algorithm: A novel approach for multiobjective optimization. Eng. Optim., 46: 12221237. CrossRef  Direct Link 
45: Yang, X.S., 2010. A New Metaheuristic BatInspired Algorithm. In: Nature Inspired Cooperative Strategies for Optimization, Gonzalez, J.R., D.A. Pelta, C. Cruz, G. Terrazas and N. Krasnogor (Eds.). Springer, Berlin, Germany, ISBN: 9783642125379, pp: 6574.
46: Yang, X.S. and S. Deb, 2009. Cuckoo search via Levy flights. Proceedings of the World Congress on Nature and Biologically Inspired Computing, December 911, 2009, Coimbatore, India, pp: 210214.
47: Yang, X.S. and A.H. Gandomi, 2012. Bat algorithm: A novel approach for global engineering optimization. Eng. Comput., 29: 464483. CrossRef  Direct Link 
48: Yang, X.S., A.H. Gandomi, S. Talatahari and A.H. Alavi, 2012. Metaheuristics in Water, Geotechnical and Transport Engineering. Newnes, Waltham, MA., ISBN13: 9780123983176, Pages: 496.
49: Zhang, Y., D. Huang, M. Ji and F. Xie, 2011. Image segmentation using PSO and PCM with mahalanobis distance. Expert Syst. Applic., 38: 90369040. CrossRef  Direct Link 
50: Mousavi, S.M., A.H. Alavi, A. Mollahasani, A.H. Gandomi and M.A. Esmaeili, 2013. Formulation of soil angle of shearing resistance using a hybrid GP and OLS method. Eng. Comput., 29: 3753. CrossRef  Direct Link 
51: Zhang, Y. and L. Wu, 2012. Artificial bee colony for two dimensional protein folding. Adv. Electrical Eng. Syst., 1: 1923. Direct Link 
52: Zhisheng, Z., 2013. Chaotic electromagnetismlike mechanism algorithm for economic load dispatch of power system. J. Exp. Theoret. Artif. Intell., 25: 493502. CrossRef  Direct Link 
53: Zhu, G. and S. Kwong, 2010. Gbestguided artificial bee colony algorithm for numerical function optimization. Applied Math. Comput., 217: 31663173. CrossRef  Direct Link 
54: Zou, D., L. Gao, S. Li and J. Wu, 2011. An effective global harmony search algorithm for reliability problems. Expert Syst. Appli., 38: 46424648. CrossRef  Direct Link 
55: Zou, D., L. Gao, S. Li and J. Wu, 2011. Solving 01 knapsack problem by a novel global harmony search algorithm. Applied Soft Comput., 11: 15561564. CrossRef  Direct Link 
56: Zou, D., L. Gao, J. Wu, S. Li and Y. Li, 2010. A novel global harmony search algorithm for reliability problems. Comput. Ind. Eng., 58: 307316. CrossRef  Direct Link 
57: Cerny, V., 1985. Thermodynamical approach to the traveling salesman problem: An efficient simulation algorithm. J. Optimiz. Theo. Appli., 45: 4151. CrossRef  Direct Link 
58: Kirkpatrick, S., C.D. Gelatt Jr. and M.P. Vecchi, 1983. Optimization by simulated annealing. Science, 220: 671680. CrossRef  Direct Link 
59: Glover, F., 1989. Tabu searchPart I. ORSA J. Comput., 1: 190206. CrossRef  Direct Link 
60: Glover, F., 1990. Tabu searchPart II. ORSA J. Comput., 2: 432. CrossRef  Direct Link 
61: Glover, F. and M. Laguna, 1997. Tabu Search. Kluwer Academic Publishers, Norwell, MA, UK.
62: Lourenco, H.R., O. Martin and T. Stutzle, 2002. Iterated Local Search. In: Handbook of Metaheuristics, Glover, F. and G. Kochenberger (Eds.). 2nd Edn., Vol. 57, Kluwer Academic Publishers, Norwell, MA., pp: 321353.
63: Fogel, L.J., A.J. Owens and M.J. Walsh, 1966. Artificial Intelligence through Simulated Evolution. John Wiley, New York, USA.
64: Holland, J.H., 1975. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control and Artificial Intelligence. 1st Edn., University of Michigan Press, Ann Arbor, MI., USA., ISBN13: 9780472084609, Pages: 183.
65: Rechenberg, I., 1973. Evolutionsstrategie: Optimierung Technischer Systeme Nach Prinzipien der Biologischen Evolution. FrommannHolzboog, Germany, ISBN13: 9783772803734, Pages: 170.
66: Schwefel, H.P., 1981. Numerical Optimization of Computer Models. John Wiley and Sons, New York, USA., ISBN13: 9780471099888, Pages: 389.
67: Guo, L., J. Karpac, S.L. Tran and H. Jasper, 2014. PGRPSC2 promotes gut immune homeostasis to limit commensal dysbiosis and extend lifespan. Cell, 156: 109122. CrossRef  Direct Link 
68: Wang, G., L. Guo, H. Wang, H. Duan, L. Liu and J. Li, 2014. Incorporating mutation scheme into krill herd algorithm for global numerical optimization. Neural Comput. Applic., 24: 853871. CrossRef  Direct Link 
69: Dumitrescu, I. and T. Stutzle, 2004. Combination of Local Search and Exact Algorthims. In: Application of Evolutionary Computing, Raidle, G.R., J.A. Meyer, M. Middendorf, S. Cagnoni and J.J.R. Cardalda et al (Eds.). Springer, New York, USA., pp: 211223.



