Abstract
Artificial bee colony (ABC) is one of the newest additions to the class of swarm intelligence. ABC algorithm has been shown to be competitive with some other populationbased algorithms. However, there is still an insufficiency that ABC is good at exploration but poor at exploitation. To make a proper balance between these two conflictive factors, this paper proposed a novel ABC variant with a timevarying strategy where the ratio between the number of employed bees and the number of onlooker bees varies with time. The linear and nonlinear timevarying strategies can be incorporated into the basic ABC algorithm, yielding ABCLTVS and ABCNTVS algorithms, respectively. The effects of the added parameters in the two new ABC algorithms are also studied through solving some representative benchmark functions. The proposed ABC algorithm is a simple and easy modification to the structure of the basic ABC algorithm. Moreover, the proposed approach is general and can be incorporated in other ABC variants. A set of 21 benchmark functions in 30 and 50 dimensions are utilized in the experimental studies. The experimental results show the effectiveness of the proposed timevarying strategy.
1. Introduction
Swarm intelligence concerns the design of intelligent optimization algorithms by taking inspirations from the collective behavior of social insects [1]. During the past decades, swarm intelligence has shown great success for solving complicated problems. The problems to be optimized by swarm intelligence algorithms do not need to be mathematically represented as continuous, convex, and/or differentiable functions; they can be represented in any form [2]. Artificial bee colony (ABC) algorithm developed by Karaboga in 2005 is a recent addition into this category [3–5].
ABC algorithm is inspired by the intelligent behavior of honey bees seeking quality food sources [3, 6–8]. In a short span of less than 10 years, ABC has already been demonstrated as a promising technique for solving global optimization problems [9, 10]. Numerical comparisons and many engineering applications demonstrated that ABC could obtain good search results [11]. Due to its simplicity, flexibility, and outstanding performance, ABC has captured increasing interests of the swarm intelligence research community and been applied to many realworld areas, such as numerical optimization [6, 10], neural network training [7, 12], finance forecasting [13, 14], production scheduling [15–17], data clustering [18, 19], image segmentation [20, 21], service selection problem [22], and power system optimization [23].
However, like its counterpart populationbased stochastic algorithms, it still has some inherent pitfalls [24–26]. The convergence speed of ABC is slower than those of the representative populationbased stochastic algorithms, such as DE and PSO [27]. Moreover, ABC suffers from premature convergence while dealing with certain complicated problems [28]. Several ABC variants have been proposed with the aim of overcoming these pitfalls. These ABC variants can be generally categorized into three groups: (1) first, the parameters of configuration tuning [29–31]; (2) second, the ABC algorithm hybridizing with other evolutionary optimization operators to enhance performance [10, 32, 33]; (3) third, the design of new learning strategy by modifying the search equation of the basic ABC algorithm [24, 25, 34].
It is recognized that exploration and exploitation are the two most important factors affecting a populationbased optimization algorithm’s performance [6, 35]. Exploration refers to the ability of a search algorithm to investigate the unknown regions in the search space in order to have high probability to discover good promising solutions. Exploitation, on the other hand, refers to the ability to concentrate the search around a promising region in order to refine a candidate solution [36, 37]. A good populationbased optimization algorithm should properly balance these two conflictive objectives [38]. It is observed that ABC has good exploration ability but poor exploitation ability [24], which may impede ABC algorithm from proceeding towards a global optimum even though the population has not converged to a local optimum.
The colony of the ABC algorithm contains three groups of bees: employed bees, onlookers, and scouts [3]. In the employed bees phase of ABC algorithm, the algorithm focuses on the explorative search indicated by the solution updating scheme, which uses the current solution and a randomly chosen solution. A fitnessbased probabilistic selection scheme is used in the onlooker phase, which indicates the exploitation tendency of the algorithm. In the original ABC algorithm, it is assumed that half of the colony consists of the employed bees, and the rest half consists of the onlookers [3]. In other words, the ratio of the employed and onlooker bees is the same, 1 : 1.
The basic ABC algorithm is easy to tune with few parameters but lacks effective and efficient ways to control the exploration ability and exploitation ability. In the present study, we propose an easy modification to the structure of the basic ABC algorithms in an attempt to balance their exploration and exploitation abilities. Compared with the basic ABC algorithm, we added a new parameter. The core idea of the modification is to design a proper mechanism, adjusting the ratio of the employed and onlooker bees with time. The proposed algorithms are called ABC with linear versus nonlinear timevarying strategy (ABCLTVS and ABCNTVS, resp.). The objective of this development is to enhance the global exploration in the early search and to encourage the solution to converge toward the global optima at the later search.
The reminder of this paper is organized as follows: Section 2 briefly introduces the basic ABC algorithm. The proposed algorithms, ABCLTVS and ABCNTVS, are elaborated in Section 3. In Section 4, comprehensive experimental studies are conducted on benchmark functions in dimension and dimension problems to verify the effectiveness of the proposed algorithms. Finally, conclusions are presented in Section 5.
2. Basic ABC Algorithm
The ABC algorithm is a recently introduced optimization algorithm proposed by Karaboga [3, 39, 40], which is inspired by the intelligent foraging behavior of the honeybee swarm. In the ABC algorithm, there are two components: the foraging artificial bees and the food source [3, 39]. The position of a food source represents a possible solution. The nectar amount of a food source corresponds to the fitness of the associated solution. In the basic ABC algorithm, the colony of artificial bees contains three groups of bees: employed bees, onlookers, and scouts. The employed bees are responsible for searching available food sources and pass the food information to the onlooker bees [3, 40]. The onlookers select good sources from those found by the employed bees to further search the foods. When the fitness of the food sources is not improved through a predetermined number of cycles, denoted as limit, the food source is abandoned by its employed bee, and then, the employed bee becomes a scout and starts to search for a new food source in the vicinity of the hive.
The ABC algorithm consists of four phases: initialization, employed bees, onlooker bees, and scout bees [3, 39, 40]. In the initialization phase of the ABC, food source positions are randomly produced in the dimensional search space using the following equation [3, 6]:where is the th food source, ; , where denotes the number of food source; and are the lower and upper bounds for dimension , respectively; is a random number uniformly distributed within range .
After producing food sources and assigning them to the employed bees. There is only one employed bee on each food source. In the employed bee phase of ABC, each employed bee tries to find a better quality food source based on . A new trial food source, denoted as , is calculated from the equation below [3, 6]:where is a randomly generated whole number in the range ; is a random number uniformly distributed in the range ; and is the index of a randomly chosen food source satisfying . After is obtained, it will be evaluated and compared to . If the fitness of is better than that of , the bee will forget the old food source and memorize the new one. Otherwise, it keeps in her memory. After all employed bees have finished their search, they share the nectar and position information of their food sources with the onlookers.
Each onlooker bee selects a food source of an employed bee to improve it. The roulette selection mechanism is performed by using (3) [3, 6]where is the probability of food source being selected by an onlooker bee and is the fitness of . The fitness of the food sources are defined as [3, 6]where is the objective function value of . Once the onlooker has selected a food source, a new candidate food source can be obtained by (2). As in the employed bees, the greedy selection between these two food sources was performed.
If a food source, , cannot be improved for a predetermined number of cycles, referred to as limit, this food source is abandoned. Then, the scout produces a new food source randomly according to (1) to replace .
The detailed pseudocode of the ABC algorithm is shown in Algorithm 1 [3, 40].

3. The Proposed Algorithm
3.1. TimeVarying Strategy
During the process of the employed bees, each food source is assigned with one employed bee. The onlooker bees select the food source based on the fitness values, which is similar to “route wheel selection” in genetic algorithm. Due to this selection scheme, it is assured that the food sources with higher fitness value have more chance to be selected by the onlooker bees. It facilitates improving the quality of these food sources. Thus, the onlooker bees are more concentrated on the exploitation ability than the employed bees. In general, for populationbased optimization algorithms, the exploration is typically preferred at the early stages of the search but is required to gradually give ways to exploitation in order to find the optimum solution efficiently [41]. In the ABC algorithm, the population size is fixed. With a large number of employed bees and relatively a small number of onlookers at the early search, it is helpful to move around the search space and enhance the exploration. On the other hand, the small number of the employed bees and the large number of the onlookers allow the individuals to converge to the global optima in the later part of the optimization. In the basic ABC algorithm, the employed bees and onlooker bees are equal to each other; that is, the ratio between the number of the employed bees and the number of the onlookers is 1 : 1. Considering those concerns, we proposed a timevarying strategy, in which the ratio between the number of the employed bees and the number of the onlookers varies with time in this paper.
The total number of the colony size, employed bees, and onlooker denoted as , , and respectively, and it holds that . The ratio between and is denoted as , . A large value of is conducive to global exploration. Conversely, it facilitates local exploitation for finetuning a local search. Intuitively, a lineardecreasing timevarying strategy (LTVS) may contribute to balance the exploration and exploitation during the entire search. A LTVS is proposed as the follows:where and are the maximum and minimum value of , respectively; is the current number of function evaluations; and is the total number of function evaluations. The number of employed bees is set to the round value of , , where is the round function with zero decimal place. If the number of employed bees is greater than the number of food source , each employed bee is assigned to one food source randomly in the first employed bees, and each of the rest of employed bees is placed at a randomly selected food source. When is smaller than , all employed bees randomly select food sources to search.
To investigate the dynamics of population distribution in basic ABC and ABC with LVTS, we herein take a typical multimodal function, Rastrigin function, with 2D dimension as an example. Figures 1 and 2 show the population distribution of Rastrigin function observed at various search phases of basic ABC and ABC with LTVS, respectively. It is noted that the value of and is set to and in ABC with LTVS in this experiment. Figures 1 and 2 show the population distribution at , , and iteration when the 2dimensional Rastrigin function was optimized by the original ABC and ABC with LTVS, respectively. Compared with the original ABC, we can obtain that the population in ABC with LTVS distributed in a wider range of search space at a relatively small iteration and gradually gathered around the global optimum. In other words, LVTS can explore in the search space in the early search stage, while converging to the global optimum with fast speed in the later search.
(a) Iterations = 5
(b) Iterations = 10
(c) Iterations = 20
(a) Iterations = 5
(b) Iterations = 10
(c) Iterations = 20
It would be interesting to know whether the nonlinear variation of can enhance the performance of the ABC algorithm. In the preset study, we proposed a nonlineardecreasing timevarying strategy (NTVS). This strategy is given as the following equation:where is the nonlinear modulation index. Figure 3 shows typical variations of with function evaluations for different settings of . With , this strategy becomes a special case of LTVS. With , varied in a convex function manner. Compared with LTVS, it can be seen that decreases in a relativeslow speed in the early search while in a faster speed in the later search. Conversely, with , varied in a concave function manner.
3.2. Parameters Tuning of TimeVarying Strategy
In order to investigate the influence of and in LTVS, six combinations for setting the values of and are tested on relevant functions including typical unimodal functions, , , and , and typical multimodal functions, , , , , and , with dimensions. The colony size is set to , the maximum functions evaluations are , and the limit is set to [42]. Experimental results of independent trials are presented in Table 1. In Table 1, “Mean” indicates the mean of function values, “SD” stands for the standard deviation, and “Rank” refers to the performance level of the certain function in the six combinations of and . The best mean function value of six combinations is marked in boldface. From the rank results given in Table 2, it shows that the settings of and are the best choice. Therefore, the parameters of and are used in LTVS. We try to obtain better performance with the NTVS of having and kept fixed to their values obtained for the linear case. Table 3 shows the performance evaluation of the resultant system with different values of , keeping . From Tables 3 and 4, we can observe that the best result is obtained with .
4. Experimental Study
4.1. Benchmark Functions and Parameters Settings
In order to test the proposed algorithm, a diverse set of benchmark functions are used to conduct the experiments. These benchmark functions can be classified into three groups: Group 1, Group 2, and Group 3. The first five functions are unimodal functions in Group 1. The next group includes ten multimodal functions with many local optima which are used to test the global search capability in avoiding premature convergence. Note that (Rosenbrock) is a unimodal problem in D or D search space but is a multimodal problem in higher dimensions. Rotated and/or shifted functions belong to Group 3. and are rotated functions, in which the original variable is rotated by left multiplying the orthogonal matrix [43], . is used to increase the complexity of the function by changing separable functions to nonseparable functions without affecting the shape of the functions. The global optima of and are shifted to different numerical values for different dimensions , where is employed to shift the global optimal solution of the original function from the center of the search space to a new location. and are complicated which are shifted and rotated. We used to represent global optimum. In each benchmark function, and represent global optimum and the corresponding function value, respectively. The function value of the best solution found by an algorithm in a run is denoted by . The error of this run is denoted as . The parameters of all benchmark functions are described in Appendix.
To validate the effectiveness of the proposed timevarying strategy, experiments were conducted to compare ABC with ABCLTVS and ABCNTVS on benchmark functions. In order to make a fair comparison, the colony size for all algorithms was set to . are used as the stop criteria for all algorithms and it is set to be for solving D problems and for solving D problems, respectively. The parameter, limit, is set to [42]. All experiments on each benchmark function were run times independently.
4.2. Experimental Results for 30D Problems
The comparative results obtained by ABC, ABCLTVS, and ABCNTVS are presented in Table 5. The best mean results on each problem among all algorithms are given in bold. In order to determine whether the results obtained by ABCLTVS and ABCNTVS are statistically different from the results generated by ABC algorithms, a twotailed test with degrees of freedom is used at a significant level of . Values of “1,” “0,” and “−1” in columns “” and “” in Table 5, respectively, denote that ABCLTVS and ABCNTVS perform significantly better than, almost the same as, and significantly worse than ABC algorithm. In order to give visualized comparisons of the involved algorithms, the convergence graphs of the best and mean function values for each ABC algorithm regarding all benchmark functions are shown in Figures 4, 5, and 6. In these figures, each curve represents the variation of mean value of error over the for a specific ABC algorithm.
(a) Sphere
(b) Schwefel’s P2.22
(c) Elliptic
(d) Noise
(e) Zakharov
(f) Rosenbrock
(a) Rastrigin
(b) Griewank
(c) Ackley
(d) Schwefel
(e) Noncontinuous Rastrigin
(f) Levy
(g) Alpine
(h) minima
(i) Generalized penalized
(a) Rotated Rosenbrock
(b) Rotated Rastrigin
(c) Shifted Rosenbrock
(d) Shifted Rastrigin
(e) Shifted Rotated Rastrigin
(f) Shifted Rotated Griewank
For solving unimodal functions, ABCNTVS achieves the highest solution accuracy on , , , and , and ABCLTVS obtains the best solution on . ABCLTVS and ABCNTVS perform significantly better than ABC algorithm on , , and . On multimodal problems, there are many local minima and it is not easy to find the global optima. ABCLTVS and ABCNTVS are shown to offer better performance than ABC algorithm on these problems. ABCLTVS can find the best solution on , , and and ABCNTVS performs the best on , , , , and . ABC, ABCLTVS, and ABCNTVS obtain the similar performance on . The performance of ABCLTVS and ABCNTVS are significantly better than that of ABC algorithm on these multimodal problems except ABCNTVS for solving .
For most of optimization algorithms, their performance will sharply decrease when solving the shifted and/or rotated problems. Table 5 shows the results on the rotated and/or shifted functions. The results appear that these three ABC algorithms are affected. We can observe that ABCLTVS and ABCNTVS still obtain relatively good performance. ABCLTVS and ABCNTVS performs significantly better than ABC algorithms on , , and . With respect to the stability of algorithms, ABCLTVS and ABCNTVS show the good stability as compared to ABC algorithms. The standard deviations of solutions found by ABCLTVS and ABCNTVS are small for the most functions. On the whole, ABCLTVS and ABCNTVS exhibit the accurate convergence precision on almost all the benchmark, which indicates the effectiveness of the proposed timevarying strategy. Moreover, experimental results demonstrate that ABCNTVS slightly outperforms ABCLTVS.
4.3. Experimental Results for 50D Problems
The experiments conducted on problems and the results for solving unimodal, multimodal, and shift and/or rotate problems are presented in Table 6. As the convergence graphs of are similar to the problems and space limitation, they are not given. Compared with ABC algorithm, ABCLTVS and ABCNTVS can still obtain highquality solutions under problems, which can be seen from Table 6. The meaning of column “” and “” in Table 6 is the same as the Table 5. It is noted that ABCNTVS, from the mean of the results, performs worse than ABC on , but not statistically significantly. According the tests results, ABCLTVS performs significantly better than ABC algorithm on all benchmark functions except , , and , so does ABCNTVS except , , and .
The experiments conducted on problems and the results for solving unimodal, multimodal, and shift and/or rotate problems are presented in Table 6. As the convergence graphs of are similar to the problems and space limitation, they are not given. Compared with ABC algorithm, ABCLTVS and ABCNTVS can still obtain highquality solutions under problems, which can be seen from Table 6. The meaning of column “” and “” in Table 6 is the same as the Table 5. According to the tests results, ABCLTVS performs significantly better than ABC algorithm on all benchmark functions except , , and , so does ABCNTVS except , , and .
4.4. GABC with TimeVarying Strategy
Inspired by particle swarm optimization (PSO), Zhu and Kwong [24] proposed a popular ABC variant, called Gbestguided artificial bee colony (GABC). GABC incorporates the information of global best (gbest) position into (2). The experimental results have shown that GABC algorithm outperforms the basic ABC algorithm. To test the effect of the timevarying strategy, we applied the proposed LTVS and NTVS into GABC algorithm, yielding GABCLTVS and GABCNTVS algorithms, respectively. Experiments are conducted on and benchmark functions to test whether the proposed timevarying strategy is effective in GABC algorithm. The parameters setting of GABC algorithm is in accordance with the original reference except the colony size and , the settings of which is same as the previous experiment.
Tables 7 and 8 give the experimental results on and problems, respectively. The best results on each problem among these three GABC algorithms are shown in bold. In addition, the columns “” and “” in Tables 7 and 8 are used to determine the statistical significance of the difference obtained by GABC with those yielded by the GABCLTVS and GABCNTVS using twotailed tests, respectively. According to the results of tests shown in Tables 7 and 8, GABCLTVS and GABCNTVS can find more accurate solutions, which are significantly better than those of GABC on about half of all benchmark functions regardless of problem dimensions.
5. Conclusions
In order to make a balance between the exploration and exploitation in ABC algorithm, a timevarying strategy has been developed. The proposed strategy is implemented through making the ratio between the number of employed bees and the numbers of onlooker bees vary with time. We have developed the two types of timevarying strategies of LTVS and NTVS. Moreover we have examined and finetuned the parameters settings of LTVS and NTVS for better performance. Comprehensive experiments have demonstrated the effectiveness of the proposed timevarying strategy in ABC and GABC algorithms. The modifications proposed in the present work are general enough to be applied in other stateoftheart ABC variants to further improve the search performance.
Appendix
A. Benchmark Functions
A.1. Unimodal Functions
(1)Sphere function (2)Schwefel’s function P2.22(3)Elliptic function(4)Noise function(5)Zakharov function
A.2. Multimodal Functions
(6)Rosenbrock function(7)Rastrigin function(8)Griewank function(9)Ackley function(10)Schwefel function(11)Noncontinuous Rastrigin function where(12)Levy function(13)Alpine function(14) minima(15)Generalized penalized where
A.3. Shifted and Rotated Functions
(16)Rotated Rosenbrock function(17)Rotated Rastrigin function(18)Shifted Rosenbrock function(19)Shifted Rastrigin function(20)Shifted Rotated Rastrigin function(21)Shifted Rotated Griewank function
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
This work is partially supported by National Natural Science Foundation of China under Grant nos. 71240015, 71402103, and 61273367; National Science Foundation of SZU under Grant 836; the Foundation for Distinguished Young Talents in Higher Education of Guangdong, China, under Grant 2012WYM_0116; the MOE Youth Foundation Project of Humanities and Social Sciences at Universities in China under Grant 13YJC630123; The Youth Foundation Project of Humanities and Social Sciences in Shenzhen University under grant 14QNFC28; and Ningbo Science & Technology Bureau (Science and Technology Project no. 2012B10055).