TSWJ The Scientific World Journal 1537-744X 2356-6140 Hindawi Publishing Corporation 10.1155/2015/904364 904364 Research Article A Seed-Based Plant Propagation Algorithm: The Feeding Station Model http://orcid.org/0000-0002-4040-6211 Sulaiman Muhammad 1, 2 Salhi Abdellah 1 Li Xinyu 1 Department of Mathematical Sciences, University of Essex, Colchester CO4 3SQ UK essex.ac.uk 2 Department of Mathematics, Abdul Wali Khan University, Mardan, Khyber Pakhtunkhwa Pakistan awkum.edu.pk 2015 232015 2015 25 12 2014 10 02 2015 10 02 2015 232015 2015 Copyright © 2015 Muhammad Sulaiman and Abdellah Salhi. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The seasonal production of fruit and seeds is akin to opening a feeding station, such as a restaurant. Agents coming to feed on the fruit are like customers attending the restaurant; they arrive at a certain rate and get served at a certain rate following some appropriate processes. The same applies to birds and animals visiting and feeding on ripe fruit produced by plants such as the strawberry plant. This phenomenon underpins the seed dispersion of the plants. Modelling it as a queuing process results in a seed-based search/optimisation algorithm. This variant of the Plant Propagation Algorithm is described, analysed, tested on nontrivial problems, and compared with well established algorithms. The results are included.

1. Introduction

Plants have evolved a variety of ways to propagate. Propagation with seeds is perhaps the most common of them all and one which takes advantage of all sorts of agents ranging from wind to water, birds, and animals. In  a Plant Propagation Algorithm based on the way the strawberry plant propagates using runners has been introduced. Here, we consider the case where the strawberry plant uses seeds to propagate.

Plants rely heavily on the dispersion of their seeds to colonise new territories and to improve their survival [2, 3]. There are a lot of studies and models of seed dispersion particularly for trees . Dispersion by wind and ballistic means is probably the most studied of all approaches . However, in the case of the strawberry plant, given the way the seeds stick to the surface of the fruit (Figure 1(a)) , dispersion by wind or mechanical means is very limited. Animals, however, and birds in particular are the ideal agents for dispersion [2, 3, 11, 12] in this case.

Strawberry plant propagation: through seed dispersion .

Strawberry fruit with seeds

Strawberry flower

A strawberry eaten by bird(s)

A bird eating strawberries

Strawberry plants showing runners and seeded fruit

There are many biologically inspired optimization algorithms in the literature [13, 14]. The Flower Pollination Algorithm (FPA) is inspired by the pollination of flowers through different agents ; the swarm data clustering algorithm is inspired by pollination by bees ; Particle Swarm Optimization (PSO) is inspired by the foraging behavior of groups of animals and insects [16, 17]; the Artificial Bee Colony (ABC) simulates the foraging behavior of honey bees [18, 19]; the Firefly algorithm is inspired by the flashing fireflies when trying to attract a mate [20, 21]; the Social Spider Optimization (SSO) algorithm is inspired by the cooperative behavior of social spiders . The list could easily be extended.

The Plant Propagation Algorithm (PPA) also known as the strawberry algorithm was inspired by the way plants and specifically the strawberry plants propagate using runners [1, 23]. The attraction of PPA is that it can be implemented easily for all sorts of optimization problems. Moreover, it has few algorithm specific arbitrary parameters. It follows the principle that plants in good spots with plenty of nutrients will send many short runners. They send few long runners when in nutrient poor spots. With long runners PPA tries to explore the search space while short runners enable it to exploit the solution space well. In this paper, we investigate an alternative PPA which is entirely based on the propagation by seeds of the strawberry plant. Because of the periodic nature of fruit and seed production, it amounts to setting up a feeding station for the attention of potential seed-dispersing agents , Hence the feeding station model used here and the resulting Seed-Based Plant Propagation Algorithm or SbPPA.

SbPPA is tested on both unconstrained and constrained benchmark problems also used in [22, 29, 30]. Experimental results are presented in Tables 47 in terms of best, mean, worst, and standard deviation for all algorithms. The paper is organised as follows. In Section 2 we briefly introduce the feeding station model representing strawberry plants in fruit and the main characteristics of the paths followed by different agents that disperse the seeds. Section 3 presents the SbPPA in pseudocode form. The experimental settings, results, and convergence graphs for different problems are given in Section 4.

2. Aspects of the Feeding Station Model

Some animals and plants depend on each other to conserve their species . Thus, many plants require, for effective seed dispersal, the visits of frugivorous birds or animals according to a certain distribution [2, 3, 32, 33].

Seed dispersal by different agents is also called “seed shadow” ; this shows the abundance of seeds spread globally or locally around parent plants. Here a queuing model is used which, in the context of a strawberry feeding station model, involves two parts:

the quantity of fruit or seeds available to agents which implies the rate at which the agents will visit the plants,

a probability density function that tells us about the service rate with which the agents are served by the plants.

The model estimates the quantity of seeds that is spread locally compared to that dispersed globally . There are two aspects that need to be balanced: exploitation, which is represented by the dispersal of seeds around the plants, and exploration which ensures that the search space is well covered.

Agents arrive at plants in a random process. Assume that at most one agent arrives to the plants in any unit of time (orderliness condition). It is further supposed that the probability of arrivals of agents to the plants remains the same for a particular period of time. This period corresponds to when the plants are in fruit and during which time the number of visitors is stable (stationarity condition). Furthermore, it is assumed that the arrival of one agent does not affect the rest of arrivals (independence).

With these assumptions in mind, the arrival of agents to plants follows a Poisson process [39, 40], which can be formally described as follows. Let X be the random variable representing the number of arrivals per unit of time t. Then, the probability of k arrivals over t is (1)PX=k=λtke-λtk!, where λ denotes the mean arrival rate of agents per time unit t. On the other hand, the time taken by agents in successfully eating fruit and leaving to disperse its seeds, in other words the service time for agents, is expressed by a random variable which follows the exponential probability distribution . This can be expressed as follows: (2)St=μe-μt, where μ is the average number of agents that can feed at time t. Let us assume that the arrival rate of agents is less than the fruits available on all plants per unit of time; therefore λ<μ.

We assume that the system is in steady state. Let A denote the average number of agents in the strawberry field (some already eating and the rest waiting to feed) and Aq the average number of agents waiting to get the chance to feed. If we denote the average number of agents eating fruits by λ/μ, then by Little’s formula , we have (3)A=Aq+λμ.

Since the plant needs to maximise dispersion, this is equivalent to having a large Aq in (3). Therefore, from this equation, we need to solve the following problem: (4)MaximizeAq=A-λμ,subject  tog1(λ,μ)=λ<μ+1,hhhhhhihhhhIhhhλ>0,μ>0, where A=10, which represents the population size in the implementation. The simple limits on the variables are 0<λ, μ100. The optimum solution to this particular problem is λ=1.1, μ=0.1, and Aq=1.

Frugivores may travel far away from the plants and hence will disperse the seeds far and wide. This feeding behaviour typically follows a Lévy distribution . In the following we present some basic facts about it.

2.1. Lévy Distribution

The Lévy distribution is a probability density distribution for random variables. Here the random variables represent the directions of flights of arbitrary birds. This function ranges over real numbers in the domain represented by the problem search space.

The flight lengths of the agents served by the plants follow a heavy tailed power law distribution , represented by (5)L(s)~s-1-β, where L(s) denotes the Lévy distribution with index β(0,2). Lévy flights are unique arbitrary excursions whose step lengths are drawn from (5). An alternative form of Lévy distribution is  (6)L(s,γ,μ)=γ2π1s-μ3/2·exp-γ2s-μ,0<μ<s<,0,Otherwise. This implies that (7)limsLs,γ,μγ2π1s3/2. In terms of the Fourier transform , the limiting value of L(s) can be written as (8)limsL(s)=αβΓ(β)sin(πβ/2)πs1+β, where Γ(β) is the Gamma function , defined by (9)Γβ=0xβ-1e-xdx. The steps L(s) are generated by Mantegna’s algorithm . This algorithm ensures that the behaviour of Lévy flights is symmetric and stable as shown in Figure 3(b).

3. Strawberry Plant Propagation Algorithm: The Feeding Station Model

We assume that the arrival of different agents (birds and animals) to the plants to feed is according to the Poisson distribution . As per the solution of problem (4), the mean arrival rate is λ=1.1, and NP=10 is the size of the agents population. Let k=1,2,,A be the possible numbers of agents visiting the plants per unit time. With these assumptions the graphic representation of (1) results in Figure 2.

Distribution of agents arriving at strawberry plants to eat fruit and disperse seeds.

Overall performance of SbPPA on Spring Design Problem.

Perturbations by  (10)

Perturbations by  (11)

As already stated, it is essential in this algorithm to balance exploration and exploitation. To this end, we choose a threshold value of the Poisson probability that dictates how much exploration and exploitation are done during the search. The probability Poiss(λ)<0.05 means that exploitation is covered. In this case, (10) below is used, which helps the algorithm to search locally: (10)xi,j*=xi,j+ξjxi,j-xl,jifPR0.8;j=1,2,,n;iii,l=1,2,,NP;il,xi,j,Otherwise, where PR denotes the rate of dispersion of the seeds locally, around SP; xi,j*  and  xi,j[ajbj] are the jth coordinates of the seeds Xi* and Xi, respectively; aj and bj are the jth lower and upper bounds defining the search space of the problem and ξj[-11]. The indices l  and  i are mutually exclusive.

On the other hand, if Poiss(λ)0.05 then global dispersion of seeds becomes more prominent. This is implemented by using the following equation: (11)xi,j*=xi,j+Lixi,j-θjifPR0.8,θjajbj,hhi=1,2,,NP;hhj=1,2,,n,xi,j,Otherwise, where Li is a step drawn from the Lévy distribution  and θj is a random coordinate within the search space. Equations (10) and (11) perturb the current solution, the results of which can be seen in Figures 3(a) and 3(b), respectively.

As mentioned in Algorithm 1, we first collect the best solutions from the first NP trial runs to form a population of potentially good solutions denoted by popbest. The convergence rate of SbPPA is shown in Figures 4 and 5 for different test problems used in our experiments (see Appendices). The statistics values best, worst, mean, and standard deviation are calculated based on popbest.

<bold>Algorithm 1: </bold>Seed-based Plant Propagation Algorithm (SbPPA) [<xref ref-type="bibr" rid="B46">47</xref>].

(1) NP Population size, r Counter of trial runs, MaxExp Maximum experiments

(2) for  r=1: MaxExp do

(3)  if  rNP  then

(4)    Create a random population of seeds pop={Xii=1,2,,NP},

using (12) and collect the best solutions from each trial run, in popbest.

(5)    Evaluate the population pop.

(6)   end if

(7)   while  r>NP  do

(8)    Use updated population popbest.

(9)   end while

(10)  while (the stopping criteria is not satisfied) do

(11)   for  i=1 to NP  do

(12)    if  Poiss(λ)i0.05, then,             (Global or local seed dispersion)

(13)     for  j=1 to n  do             (n is number of dimensions)

(14)      if  rand PR  then,              (PR = Perturbation Rate)

(15)       Update the current entry according to (11)

(16)      end if

(17)     end for

(18)    else

(19)     for  j=1 to n  do

(20)      if  rand PR  then,

(21)       Update the current entry according to (10)

(22)       end if

(23)     end for

(24)    end if

(25)   end for

(26)   Update current best

(27)  end while

(28)  Return: Updated population and global best solution.

(29) end for

Performance of SbPPA on unconstrained global optimization problems.

Performance of SbPPA on constrained global optimization problems (see Appendices).

The seed-based propagation process of SP can be represented in the following steps.

The dispersal of seeds in the neighbourhood of the SP, as shown in Figure 1(e), is carried out either by fruits fallen from strawberry plants after they become ripe or by agents. The step lengths for this phase are calculated using (10).

Seeds are spread globally through agents, as shown in Figures 1(c) and 1(d). The step lengths for these travelling agents are drawn from the Lévy distribution .

The probabilities, Poiss(λ), that a certain number k of agents will arrive to SP to eat fruits and disperse it, is used as a balancing factor between exploration and exploitation.

For implementation purposes, we assume that each SP produces one fruit, and each fruit is assumed to have one seed; by a solution Xi we mean the current position of the ith seed to be dispersed. The number of seeds in the population is denoted by NP. Initially we generate a random population of NP seeds using (12)xi,j=aj+bj-ajηj,j=1,,n, where xi,j[ajbj] is the jth coordinate of solution Xi, aj and bj are the jth coordinates of the bounds describing the search space of the problem, and ηj(01). This means that Xi=[xi,j], forj=1,,n, represents the position of the jth seed in population pop.

4. Experimental Settings and Discussion

In our experiments we tested SbPPA against some recently developed algorithms and some well established and standard ones. Our set of test problems includes benchmark constrained and unconstrained optimization problems [22, 30, 48, 49]. The results are compared in terms of statistics (best, worst, mean and standard deviation) for solutions obtained by SbPPA; ABC [18, 50]; PSO ; FF ; HPA ; SSO-C ; Classical Evolutionary Programming (CEP) ; and Fast Evolutionary Programming (FEP) . The detailed descriptions of these problems are given in Appendices.

In Tables 4 and 7, the significance of results is shown in terms of win/tie/loss (see Table  2 in ) according to the following notations:

(+) when SbPPA is better;

() when the results are approximately the same as those obtained with SbPPA;

(−) when SbPPA is worse.

Moreover, in Tables 5 and 6 the significance of results obtained with SbPPA is highlighted.

4.1. Parameter Settings

The parameter settings are given in Tables 13.

Parameters used for each algorithm for solving unconstrained global optimization problems f1f10. All experiments are repeated 30 times.

PSO [16, 29] ABC [18, 29] HPA  SbPPA 
M = 100 SN = 100 Agents = 100 NP = 10
G max = ( Dimension × 20,000 ) M MCN = (Dimension×20,000)SN Iteration number = (Dimension×20,000)Agents Iteration number = (Dimension×20,000)NP
c 1 = 2 MR = 0.8 c 1 = 2 PR = 0.8
c 2 = 2 Limit = (SN×dimension)2 c 2 = 2 Poiss ( λ ) = 0.05
W = (Gmax-iterationindex)Gmax Limit = (Agents×dimension)2
W = (Iterationnumber-iterationindex)Iterationnumber

Experimental setup used for each algorithm for solving unconstrained global optimization problems f11f26. All experiments are repeated 50 times.

CEP [30, 53, 54] FEP  SbPPA
Population size μ=100 Population size μ=100 NP = 10
Tournament size q=10 Tournament size q = 10 PR = 0.8
η = 3.0 η = 3.0 Poiss ( λ ) = 0.05

Parameters used for each algorithm for solving constrained optimization problems. All experiments are repeated 30 times.

PSO  ABC  FF  SSO-C  SbPPA 
M = 250 SN = 40 Fireflies = 25 N = 50 NP = 10
G max = 300 MCN = 6000 Iteration number = 2000 Iteration number = 500 Iteration number = 800
c 1 = 2 MR = 0.8 q = 1.5 PF = 0.7 PR = 0.8
c 2 = 2 α = 0.001 Poiss ( λ ) = 0.05
Weight factors = 0.9 to 0.4

Results obtained by SbPPA, HPA, PSO, and ABC. All problems in this table are unconstrained.

Fun. Dim Algorithm Best Worst Mean SD
f 1 4 ABC ( + ) 0.0129 ( + ) 0.6106 ( + ) 0.1157 ( + ) 0.111
PSO ( - ) 6.8991 E - 08 ( + ) 0.0045 ( + ) 0.001 ( + ) 0.0013
HPA ( + ) 2.0323 E - 06 ( + ) 0.0456 ( + ) 0.009 ( + ) 0.0122
SbPPA 1.08 E - 07 7.05 E - 06 3.05 E - 06 3.14 E - 06

f 2 2 ABC ( + ) 1.2452 E - 08 ( + ) 8.4415 E - 06 ( + ) 1.8978 E - 06 ( + ) 1.8537 E - 06
PSO ( ) 0 ( ) 0 ( ) 0 ( ) 0
HPA ( ) 0 ( ) 0 ( ) 0 ( ) 0
SbPPA 0 0 0 0

f 3 2 ABC ( ) 0 ( + ) 4.8555 E - 06 ( + ) 4.1307 E - 07 ( + ) 1.2260 E - 06
PSO ( ) 0 ( + ) 3.5733 E - 07 ( + ) 1.1911 E - 08 ( + ) 6.4142 E - 08
HPA ( ) 0 ( ) 0 ( ) 0 ( ) 0
SbPPA 0 0 0 0

f 4 2 ABC ( ) - 1.03163 ( ) - 1.03163 ( ) - 1.03163 ( ) 0
PSO ( ) - 1.03163 ( ) - 1.03163 ( ) - 1.03163 ( ) 0
HPA ( ) - 1.03163 ( ) - 1.03163 ( ) - 1.03163 ( ) 0
SbPPA −1.031628 −1.031628 −1.031628 0

f 5 6 ABC ( ) - 50.0000 ( ) - 50.0000 ( ) - 50.0000 ( - ) 0
PSO ( ) - 50.0000 ( ) - 50.0000 ( ) - 50.0000 ( - ) 0
HPA ( ) - 50.0000 ( ) - 50.0000 ( ) - 50.0000 ( - ) 0
SbPPA - 50.0000 - 50.0000 - 50.0000 5.88 E - 09

f 6 10 ABC ( + ) - 209.9929 ( + ) - 209.8437 ( + ) - 209.9471 ( + ) 0.044
PSO ( ) - 210.0000 ( ) - 210.0000 ( ) - 210.0000 ( - ) 0
HPA ( ) - 210.0000 ( ) - 210.0000 ( ) - 210.0000 ( + ) 1
SbPPA - 210.0000 - 210.0000 - 210.0000 4.86 E - 06

f 7 30 ABC ( + ) 2.6055 E - 16 ( + ) 5.5392 E - 16 ( + ) 4.7403 E - 16 ( + ) 9.2969 E - 17
PSO ( ) 0 ( ) 0 ( ) 0 ( ) 0
HPA ( ) 0 ( ) 0 ( ) 0 ( ) 0
SbPPA 0 0 0 0

f 8 30 ABC ( + ) 2.9407 E - 16 ( + ) 5.5463 E - 16 ( + ) 4.8909 E - 16 ( + ) 9.0442 E - 17
PSO ( ) 0 ( ) 0 ( ) 0 ( ) 0
HPA ( ) 0 ( ) 0 ( ) 0 ( ) 0
SbPPA 0 0 0 0

f 9 30 ABC ( ) 0 ( + ) 1.1102 E - 16 ( + ) 9.2519 E - 17 ( + ) 4.1376 E - 17
PSO ( ) 0 ( + ) 1.1765 E - 01 ( + ) 2.0633 E - 02 ( + ) 2.3206 E - 02
HPA ( ) 0 ( ) 0 ( ) 0 ( ) 0
SbPPA 0 0 0 0

f 10 30 ABC ( + ) 2.9310 E - 14 ( + ) 3.9968 E - 14 ( + ) 3.2744 E - 14 ( + ) 2.5094 E - 15
PSO ( ) 7.9936 E - 15 ( + ) 1.5099 E - 14 ( + ) 8.5857 E - 15 ( + ) 1.8536 E - 15
HPA ( ) 7.9936 E - 15 ( + ) 1.5099 E - 14 ( + ) 1.1309 E - 14 ( + ) 3.54 E - 15
SbPPA 7.994 E - 15 7.99361 E - 15 7.994 E - 15 7.99361 E - 15

Results obtained by SbPPA, CEP, and FEP. All problems in this table are unconstrained .

Function number Algorithm Maximum generations Mean SD
f 11 CEP 2000 2.60 E - 03 1.70 E - 04
FEP 8.10 E - 03 7.70 E - 04
SbPPA 9.45 E - 13 4.08 E - 12

f 12 CEP 5000 2 1.2
FEP 0.3 0.5
SbPPA 3.93 E - 02 3.76 E - 02

f 13 CEP 20000 6.17 13.61
FEP 5.06 5.87
SbPPA 1.86 E + 01 2.25 E + 00

f 14 CEP 1500 577.76 1125.76
FEP 0 0
SbPPA 0 0

f 15 CEP 3000 1.80 E - 02 6.40 E - 03
FEP 7.60 E - 03 2.60 E - 03
SbPPA 3.61 E - 03 1.31 E - 03

f 16 CEP 9000 - 7.92 E + 03 6.35 E + 02
FEP - 1.26 E + 04 5.26 E + 01
SbPPA - 1.16 E + 04 6.04 E + 01

f 17 CEP 5000 89 23.1
FEP 4.60 E - 02 1.20 E - 02
SbPPA 8.73 E + 00 9.88 E - 01

f 18 CEP 100 0.398 1.50 E - 07
FEP 0.398 1.50 E - 07
SbPPA 3.98 E - 01 0

Results obtained by SbPPA, CEP, and FEP. All problems in this table are unconstrained .

Function number Algorithm Maximum generations Mean SD
f 19 CEP 100 3 0
FEP 3.02 0.11
SbPPA 3 3.05 E - 15

f 20 CEP 100 - 3.86 E + 00 1.40 E - 02
FEP - 3.86 E + 00 1.40 E - 05
SbPPA - 3.86 E + 00 2.75 E - 15

f 21 CEP 200 - 3.28 E + 00 5.80 E - 02
FEP - 3.27 E + 00 5.90 E - 02
SbPPA - 3.32 E + 00 2.91 E - 14

f 22 CEP 100 - 6.86 E + 00 2.67 E + 00
FEP - 5.52 E + 00 1.59 E + 00
SbPPA - 1.02 E + 01 4.30 E - 09

f 23 CEP 100 - 8.27 2.95
FEP - 5.52 2.12
SbPPA - 1.04 E + 01 7.73 E - 09

f 24 CEP 100 - 9.1 2.92
FEP - 6.57 3.14
SbPPA - 1.05 E + 01 1.03 E - 07

f 25 CEP 100 1.66 1.19
FEP 1.22 0.56
SbPPA 9.98 E - 01 1.13 E - 16

f 26 CEP 4000 4.70 E - 04 3.00 E - 04
FEP 5.00 E - 04 3.20 E - 04
SbPPA 3.07 E - 04 6.80 E - 15

Results obtained by SbPPA, PSO, ABC, FF, and SSO-C. All problems in this table are standard constrained optimization problems.

Fun. name Optimal Algorithm Best Mean Worst SD
CP1 −15 PSO ( ) - 15 ( ) - 15 ( ) - 15 ( - ) 0
ABC ( ) - 15 ( ) - 15 ( ) - 15 ( - ) 0
FF ( + ) 14.999 ( + ) 14.988 ( + ) 14.798 ( + ) 6.40 E - 07
SSO-C ( ) - 15 ( ) - 15 ( ) - 15 ( - ) 0
SbPPA - 15 - 15 - 15 1.95 E - 15

CP2 - 30665.539 PSO ( ) - 30665.5 ( + ) - 30662.8 ( + ) - 30650.4 ( + ) 5.20 E - 02
ABC ( ) - 30665.5 ( + ) - 30664.9 ( + ) - 30659.1 ( + ) 8.20 E - 02
FF ( ) - 3.07 E + 04 ( + ) - 30662 ( + ) - 30649 ( + ) 5.20 E - 02
SSO-C ( ) - 3.07 E + 04 ( ) - 30665.5 ( + ) - 30665.1 ( + ) 1.10 E - 04
SbPPA - 30665.5 - 30665.5 - 30665.5 2.21 E - 06

CP3 - 6961.814 PSO ( + ) - 6.96 E + 03 ( + ) - 6958.37 ( + ) - 6942.09 ( + ) 6.70 E - 02
ABC ( - ) - 6961.81 ( + ) - 6958.02 ( + ) - 6955.34 ( - ) 2.10 E - 02
FF ( + ) - 6959.99 ( + ) - 6.95 E + 03 ( + ) - 6947.63 ( - ) 3.80 E - 02
SSO-C ( - ) - 6961.81 ( + ) - 6961.01 ( + ) - 6960.92 ( - ) 1.10 E - 03
SbPPA - 6961.5 - 6961.38 - 6961.45 0.043637

CP4 24.306 PSO ( - ) 24.327 ( + ) 2.45 E + 01 ( + ) 24.843 ( + ) 1.32 E - 01
ABC ( + ) 24.48 ( + ) 2.66 E + 01 ( + ) 28.4 ( + ) 1.14
FF ( - ) 23.97 ( + ) 28.54 ( + ) 30.14 ( + ) 2.25
SSO-C ( - ) 24.306 ( - ) 24.306 ( - ) 24.306 ( - ) 4.95 E - 05
SbPPA 24.34442 24.37536 24.37021 0.012632

CP5 - 0.7499 PSO ( ) - 0.7499 ( + ) - 0.749 ( + ) - 0.7486 ( + ) 1.20 E - 03
ABC ( ) - 0.7499 ( + ) - 0.7495 ( + ) - 0.749 ( + ) 1.67 E - 03
FF ( + ) - 0.7497 ( + ) - 0.7491 ( + ) - 0.7479 ( + ) 1.50 E - 03
SSO-C ( ) - 0.7499 ( ) - 0.7499 ( ) - 0.7499 ( - ) 4.10 E - 09
SbPPA 0.7499 0.749901 0.7499 1.66 E - 07

Spring Design Problem Not known PSO ( + ) 0.012858 ( + ) 0.014863 ( + ) 0.019145 ( + ) 0.001262
ABC ( ) 0.012665 ( + ) 0.012851 ( + ) 0.01321 ( + ) 0.000118
FF ( ) 0.012665 ( + ) 0.012931 ( + ) 0.01342 ( + ) 0.001454
SSO-C ( ) 0.012665 ( + ) 0.012765 ( + ) 0.012868 ( + ) 9.29 E - 05
SbPPA 0.012665 0.012666 0.012666 3.39 E - 10

Welded beam design problem Not known PSO ( + ) 1.846408 ( + ) 2.011146 ( + ) 2.237389 ( + ) 0.108513
ABC ( + ) 1.798173 ( + ) 2.167358 ( + ) 2.887044 ( + ) 0.254266
FF ( + ) 1.724854 ( + ) 2.197401 ( + ) 2.931001 ( + ) 0.195264
SSO-C ( ) 1.724852 ( + ) 1.746462 ( + ) 1.799332 ( + ) 0.02573
SbPPA 1.724852 1.724852 1.724852 4.06 E - 08

Speed reducer design optimization Not known PSO ( + ) 3044.453 ( + ) 3079.262 ( + ) 3177.515 ( + ) 26.21731
ABC ( + ) 2996.116 ( + ) 2998.063 ( + ) 3002.756 ( + ) 6.354562
FF ( + ) 2996.947 ( + ) 3000.005 ( + ) 3005.836 ( + ) 8.356535
SSO-C ( ) 2996.113 ( ) 2996.113 ( ) 2996.113 ( + ) 1.34 E - 12
SbPPA 2996.114 2996.114 2996.114 0

5. Conclusion

In this paper, a new metaheuristic referred to as the Seed-Based Plant Propagation Algorithm (SbPPA)  has been proposed. Plants have evolved a variety of ways to propagate. Propagation through seeds is perhaps the most common of them all and one which takes advantage of all sorts of agents ranging from wind to water, birds, and animals. The strawberry plant uses both runners and seeds to propagate. Here we consider the propagation through seeds that the strawberry plant has evolved, to design an efficient optimization algorithm.

To capture the dispersal process, we adopt a queuing approach which, given the extent of fruit produced, indicates the extent of seeds dispersed and hence the effectiveness of the search/optimization algorithm based on this process. Looking at the random process of agents using the plants (feeding station) it is reasonable to assume that it is of the Poisson type. On the other hand, the time taken by agents in successfully eating fruit and leaving to disperse its seeds, in other words the service time for agents, is expressed by a random variable which follows the exponential probability distribution. To this end, we choose a threshold value of the Poisson probability that dictates how much exploration and exploitation are done during the search. An alternative strategy has been adopted here. This strategy consists in making sure that the initial population is as good as the user can afford it to be by using best solutions found so far. The effects of this strategy on convergence are shown through convergence plots of Figures 4 and 5, for some of the solved problems. SbPPA is easy to implement as it requires less arbitrary parameter settings than other algorithms. The success rate of SbPPA increases as it gets its population of best solutions. It has been implemented for both unconstrained and constrained optimization problems. Its performance, compared to that of other algorithms, points to SbPPA as being superior.

Appendices A. Unconstrained Global Optimization Problems

See Tables 8 and 9.

Unconstrained global optimization problems (Set-1) used in our experiments.

Fun. Fun. name D C Range Min Formulation
f 1 Colville 4 UN - 10 10 D 0 f x = 100 x 1 2 - x 2 + x 1 - 1 2 + x 3 - 1 2 + 90 x 3 2 - x 4 2
+ 10.1 ( x 2 - 1 2 + ( x 4 - 1 ) 2 ) + 19.8 x 2 - 1 x 4 - 1

f 2 Matyas 2 UN - 10 10 D 0 f ( x ) = 0.26 x 1 2 + x 2 2 - 0.48 x 1 x 2

f 3 Schaffer 2 MN - 100 100 D 0 f ( x ) = 0.5 + sin 2 i = 1 n x i 2 - 0.5 1 + 0.001 i = 1 n x i 2 2

f 4 Six Hump Camel Back 2 MN - 5 5 D - 1.03163 f ( x ) = 4 x 1 2 - 2.1 x 1 4 + 1 3 x 1 6 + x 1 x 2 - 4 x 2 2 + 4 x 2 4

f 5 Trid6 6 UN - 36 36 D - 50 f x = i = 1 6 x i - 1 2 - i = 2 6 x i x i - 1

f 6 Trid10 10 UN - 100 100 D - 210 f ( x ) = i = 1 10 x i - 1 2 - i = 2 10 x i x i - 1

f 7 Sphere 30 US - 100 100 D 0 f ( x ) = i = 1 n x i 2

f 8 SumSquares 30 US - 10 10 D 0 f ( x ) = i = 1 n i x i 2

f 9 Griewank 30 MN - 600 600 D 0 f ( x ) = 1 4000 i = 1 n x i 2 - i = 1 n cos x i i + 1

f 10 Ackley 30 MN - 32 32 D 0 f ( x ) = - 20 exp - 0.2 1 n i = 1 n x i 2 - exp 1 n i = 1 n cos 2 π x i + 20 + e

Unconstrained global optimization problems (Set-2) used in our experiments .

 Fun. number Range D Function Formulation f min ⁡ f 11 [ −10, 10]D 30 Schwefel's Problem 2.22 f x = ∑ i = 1 n ‍ | x i | + ∏ i = 1 n | x i | 0 f 12 [ −100, 100]D 30 Schwefel's Problem 2.21 f x = ma x i x i , 1 ≤ i ≤ n 0 f 13 [ −10, 10]D 30 Rosenbrock f x = ∑ i = 1 n - 1 ‍ 100 x i + 1 - x i 2 2 + x i - 1 2 0 f 14 [ −100, 100]D 30 Step f ( x ) = ∑ i = 1 n ‍ x i + 0.5 2 0 f 15 [ −1.28, 1.28]D 30 Quartic (noise) f ( x ) = ∑ i = 1 n ‍ i x i 4 + random [ 0,1 ) 0 f 16 [ −500, 500]D 30 Schwefel f ( x ) = - ∑ i = 1 n ‍ x i sin ⁡ x i - 12569.5 f 17 [ −5.12, 5.12]D 30 Rastrigin f ( x ) = x i 2 - 10cos 2 π x i + 10 0 f 18 [ −5, 10] × [0, 15] 2 Branin f ( x ) = x 2 - 5.1 4 π 2 x 1 2 + 5 π x 1 - 6 2 + 10 1 - 1 8 π cos ⁡ x 1 + 10 0.398 f 19 [ −2, 2]D 2 Goldstein-Price f x = 1 + ( x 1 + x 2 + 1 ) 2 19 - 14 x 1 + 3 x 1 2 - 14 x 2 + 6 x 1 x 2 + 3 x 2 2 f x = ×   30 + 2 x 1 - 3 x 2 2 18 - 32 x 1 + 12 x 1 2 + 48 x 2 - 36 x 1 x 2 + 27 x 2 2 3 f 20 [ 0, 1]D 4 Hartman's Family (n=3) f x = - ∑ i = 1 4 ‍ c i exp ⁡ ∑ j = 1 3 ‍ a i j ( x j - p i j ) 2 - 3.86 f 21 [ 0, 1]D 6 Hartman's Family (n=6) f ( x ) = - ∑ i = 1 4 ‍ c i exp ⁡ ∑ j = 1 6 ‍ a i j ( x j - p i j ) 2 - 3.32 f 22 [ 0, 10]D 4 Shekel's Family (m=5) f ( x ) = - ∑ i = 1 5 ‍ ( x - a i ) ( x - a i ) T + c i - 1 - 10 f 23 [ 0, 10]D 4 Shekel's Family (m=7) f ( x ) = - ∑ i = 1 7 ‍ x - a i x - a i T + c i - 1 - 10 f 24 [ 0, 10]D 4 Shekel's Family (m=10) f ( x ) = - ∑ i = 1 10 ‍ x - a i x - a i T + c i - 1 - 10 f 25 [ −65.536, 65.536]D 2 Shekel's Foxholes f ( x ) = 1 500 + ∑ j = 1 25 ‍ 1 j + ∑ i = 1 2 ‍ x i - a i j 6 - 1 1 f 26 [ −5, 5]D 4 Kowalik f ( x ) = ∑ i = 1 11 ‍ a i - x 1 b i 2 + b i x 2 b i 2 + b i x 3 + x 4 2 0.0003075

B. Set of Constrained Global Optimization Problems Used in Our Experiments B.1. CP1

Consider the following: (B.1)Minf(x)=5d=14xd-5d=14xd2-d=513xd,subject  tog1(x)=2x1+2x2+x10+x11-100,hhhhhhhhg2(x)=2x1+2x3+x10+x12-100,hhhhhhhhg3(x)=2x2+2x3+x11+x12-100,hhhhhhhhg4(x)=-8x1+x100,hhhhhhhhg5(x)=-8x2+x110,hhhhhhhhg6(x)=-8x3+x120,hhhhhhhhg7(x)=-2x4-x5+x100,hhhhhhhhg8(x)=-2x6-x7+x110,hhhhhhhhg9(x)=-2x8-x9+x120, where bounds are 0xi1(i=1,,9,13), 0xi100(i=10,11,12). The global optimum is at x*=(1,1,1,1,1,1,1,1,1,,3,3,3,1), f(x*)=-15.

B.2. CP2

Consider the following: (B.2)Minf(x)=5.3578547x2+0.8356891x1x5hhhhhhhhhhhihihh+37.293239x1-40792.141,subject  tog1(x)=85.334407+0.0056858x2x5hhhhhhhhhhhhhhhh+0.0006262x1x4-0.0022053x3x5hhhhhhhhhhhhhhhh-920,hhhhhhhhg2(x)=-85.334407-0.0056858x2x5hhhhhhhhhhhhhhhh-0.0006262x1x4+0.0022053x3x5hhhhhhhhhhhhhh0,hhhhhhhhg3(x)=80.51249+0.0071317x2x5hhhhhhhhhhhhhhhh+0.0029955x1x2-0.0021813x2hhhhhhhhhhhhhhhh-1100,hhhhhhhhg4(x)=-80.51249-0.0071317x2x5hhhhhhhhhhhhhhhh+0.0029955x1x2-0.0021813x2hhhhhhhhhhhhhhhh+900,hhhhhhhhg5(x)=9.300961-0.0047026x3x5hhhhhhhhhhhhhhhh-0.0012547x1x3-0.0019085x3x4hhhhhhhhhhhhhhhh-250,hhhhhhhhg6(x)=-9.300961-0.0047026x3x5hhhhhhhhhhhhhhhh-0.0012547x1x3-0.0019085x3x4hhhhhhhhhhhhhhhh+200, where 78x1102, 33x245, and 27xi45(i=3,4,5). The optimum solution is x*=(78,33,29.995256025682,45,36.775812905788), where f(x*)=-30665.539. Constraints g1 and g6 are active.

B.3. CP3

Consider the following: (B.3)Minf(x)=x1-103+x2-203,subject  tog1(x)=-x1-52-x2-52+1000,hhhhhhhhg2(x)=x1-62+x2-52-82.810, where 13x1100 and 0x2100. The optimum solution is x*=(14.095,0.84296) where f(x*)=-6961.81388. Both constraints are active.

B.4. CP4

Consider the following: (B.4)Minfx=x12+x22+x1x2-14x1-16x2hhhhhhhhhihhhh+x3-102+4x4-52+x5-32hhhhhhhhhihhhh+2x6-12+5x72+7x8-112hhhhhhhhhihhhh+2x9-102+x10-72+45,subject  tog1x=-105+4x1+5x2-3x7+9x80,hhhhhhhhg2x=10x1-8x2-17x7+2x80,hhhhhhhhg3x=-8x1+2x2+5x9-2x10-120,hhhhhhhhg4x=3x1-22+4x2-32+2x32hhhhhhhhhhhhhhhh-7x4-1200,hhhhhhhhg5x=5x12+8x2+x3-62-2x4hhhhhhhhhhhhhhhh-400,hhhhhhhhg6x=x12+2x2-22-2x1x2+14x5hhhhhhhhhhhhhhhh-6x60,hhhhhhhhg7x=0.5x1-82+2x2-42+3x52hhhhhhhhhhhhhhhh-x6-300,hhhhhhhhg8x=-3x1+6x2+12x9-82-7x100, where -10xi10(i=1,,10). The global optimum is x*=(2.171996, 2.363683, 8.773926, 5.095984, 0.9906548, 1.430574, 1.321644, 9.828726, 8.280092,8.375927), where f(x*)=24.3062091. Constraints g1,g2,g3,g4,g5, and g6 are active.

B.5. CP5

Consider the following: (B.5)Minf(x)=x12+x2-12,subject  tog1(x)=x2-x12=0, where 1x11, 1x21. The optimum solution is x*=(±1/(2),1/2), where f(x*)=0.7499.

B.6. Welded Beam Design Optimisation

The welded beam design is a standard test problem for constrained design optimisation [55, 56]. There are four design variables: the width w and length L of the welded area and the depth d and thickness h of the main beam. The objective is to minimise the overall fabrication cost, under the appropriate constraints of shear stress τ, bending stress σ, buckling load P, and maximum end deflection δ. The optimization model is summarized as follows, where xT=w,L,d,h: (B.6)Minimisef(x)=1.10471w2L+0.04811dh(14.0+L),subject  tog1(x)=w-h0,hhhhhhhhhg2(x)=δ(x)-0.250,hhhhhhhhhg3(x)=τ(x)-13,6000,hhhhhhhhhg4(x)=σ(x)-30,0000,hhhhhhhhhg5(x)=1.10471w2+0.04811dh(14.0+L)hhhhhhhhhhhhhhh-5.00,hhhhhhhhhg6(x)=0.125-w0,hhhhhhhhhg7(x)=6000-P(x)0, where (B.7)σ(x)=504,000hd2,D=12L2+w+d2,Q=600014+L2,δ=65,85630,000hd3,J=2wLL26+w+d22,α=60002wL,β=QDJ,P=0.61423×106dh361-30/48d28,τ(x)=α2+αβLD+β2.

B.7. Speed Reducer Design Optimization

The problem of designing a speed reducer  is a standard test problem. It consists of the design variables as face width x1, module of teeth x2, number of teeth on pinion x3, length of the first shaft between bearings x4, length of the second shaft between bearings x5, diameter of the first shaft x6, and diameter of the first shaft x7 (all variables are continuous except x3 that is integer). The weight of the speed reducer is to be minimized subject to constraints on bending stress of the gear teeth, surface stress, transverse deflections of the shafts, and stresses in the shaft . The mathematical formulation of the problem, where xT=(x1,x2,x3,x4,x5,x6,x7), is as follows: (B.8)Minimisefx=0.7854x1x22hhhhhhhhhhhhhhh·(3.3333x32+14.9334x343.0934)hhhhhhhhhhhhhhh-1.508x1(x62+x73)hhhhhhhhhhhhhhh+7.4777(x63+x73)hhhhhhhhhhhhhhh+0.7854(x4x62+x5x72),subject  tog1(x)=27x1x22x3-10,hhhhhhhhhg2(x)=397.5x1x22x32-10,hhhhhhhhhg3(x)=1.93x43x2x3x64-10,hhhhhhhhhg4(x)=1.93x53x2x3x74-10,hhhhhhhhhg5(x)=1.0110x63745.0x4x2x32+16.9×106hhhhhhhhhhhhhhhi-10,hhhhhhhhhg6(x)=1.085x73745.0x5x2x32+157.5×106hhhhhhhhhhhhhhhi-10,hhhhhhhhhg7(x)=x2x340-10,hhhhhhhhhg8(x)=5x2x1-10,hhhhhhhhhg9(x)=x112x2-10,hhhhhhhhhg10(x)=1.5x6+1.9x4-10,hhhhhhhhhg11(x)=1.1x7+1.9x5-10. The simple limits on the design variables are 2.6x13.6, 0.7x20.8, 17x328, 7.3x48.3, 7.8x58.3, 2.9x63.9, and 5.0x75.5.

B.8. Spring Design Optimisation

The main objective of this problem [58, 59] is to minimize the weight of a tension/compression string, subject to constraints of minimum deflection, shear stress, surge frequency, and limits on outside diameter and on design variables. There are three design variables: the wire diameter x1, the mean coil diameter x2, and the number of active coils x3 . The mathematical formulation of this problem, where xT=(x1,x2,x3), is as follows: (B.9)Minimizef(x)=(x3+2)x2x12,subject  tog1(x)=1-x23x37,178x140,hhhhihhhhhg2(x)=4x22-x1x212,566x2x13-x14+15,108x12hhhhhhhhhhhhhhhh-10,hhhhhihhhhg3(x)=1-140.45x1x22x30,hhhhhihhhhg4(x)=x2+x11.5-10. The simple limits on the design variables are 0.05x12.0, 0.25x21.3, and 2.0x315.0.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgments

The authors are grateful to anonymous reviewers for their valuable reviews and constructive criticism on earlier version of this paper. This work is supported by Abdul Wali Khan University, Mardan, Pakistan, Grant no. F.16-5/P& D/AWKUM/238.

Salhi A. Fraga E. S. Nature-inspired optimisation approaches and the new plant propagation algorithm Proceedings of the International Conference on Numerical Analysis and Optimization (ICeMATH '11) 2011 Yogyakarta, Indonesia K2-1 K2-8 Herrera C. M. Pellmyr O. Plant Animal Interactions: An Evolutionary Approach 2009 John Wiley & Sons Herrera C. M. Seed dispersal by vertebrates Plant-Animal Interactions: An Evolutionary Approach 2002 185 208 Abrahamson W. G. Taylor T. N. Plant-Animal Interactions 1989 McGraw Hill Andersen A. N. Braithwaite R. W. Plant-animal interactions Landscape and Vegetation Ecology of the Kakadu Region, Northern Australia 1996 Springer 137 154 Bryant J. P. Plant-animal interactions Environmental Entomology 1990 19 4 1169 1170 Glover B. J. Understanding Flowers and Flowering: an Integrated Approach 2007 Oxford, UK Oxford University Press Yang X.-S. Flower pollination algorithm for global optimization Unconventional Computation and Natural Computation 2012 Springer 240 249 Yang X.-S. Karamanoglu M. He X. Multi-objective flower algorithm for optimization 18 Proceedings of the 13th Annual International Conference on Computational Science (ICCS '13) June 2013 861 868 10.1016/j.procs.2013.05.251 2-s2.0-84891688711 du Plessis H. J. Brand R. J. Glyn-Woods C. Goedhart M. A. Efficient genetic transformation of strawberry (Fragaria x Ananassa Duch.) cultivar selekta III International Symposium on In Vitro Culture and Horticultural Breeding 1996 ISHS 289 294 ISHS Acta Horticulturae 447 Krefting L. W. Roe E. I. The role of some birds and mammals in seed germination Ecological Monographs 1949 19 3 269 286 10.2307/1943538 Wenny D. G. Levey D. J. Directed seed dispersal by bellbirds in a tropical cloud forest Proceedings of the National Academy of Sciences of the United States of America 1998 95 11 6204 6207 10.1073/pnas.95.11.6204 2-s2.0-0032568585 Brownlee J. Clever Algorithms: Nature-Inspired Programming Recipes 2011 Yang X.-S. Nature-Inspired Metaheuristic Algorithms 2011 Beckington, UK Luniver Press Kazemian M. Ramezani Y. Lucas C. Moshiri B. Swarm clustering based on flowers pollination by artificial bees Studies in Computational Intelligence 2006 34 191 202 10.1007/978-3-540-34956-3_8 2-s2.0-33846286451 Eberhart R. Kennedy J. A new optimizer using particle swarm theory Proceedings of the 6th International Symposium on Micro Machine and Human Science (MHS '95) October 1995 Nagoya, Japan IEEE 39 43 Clerc M. Particle Swarm Optimization 2010 93 John Wiley & Sons Karaboga D. An idea based on honey bee swarm for numerical optimization 2005 TR06 Kayseri, Turkey Erciyes University Press Karaboga D. Basturk B. On the performance of artificial bee colony (ABC) algorithm Applied Soft Computing Journal 2008 8 1 687 697 10.1016/j.asoc.2007.05.007 2-s2.0-34548479029 Yang X.-S. Firefly algorithm, stochastic test functions and design optimisation International Journal of Bio-Inspired Computation 2010 2 2 78 84 10.1504/ijbic.2010.032124 2-s2.0-79953855364 Gandomi A. H. Yang X.-S. Alavi A. H. Mixed variable structural optimization using Firefly Algorithm Computers and Structures 2011 89 23-24 2325 2336 10.1016/j.compstruc.2011.08.002 2-s2.0-80055049702 Cuevas E. Cienfuegos M. A new algorithm inspired in the behavior of the social-spider for constrained optimization Expert Systems with Applications 2014 41 2 412 425 10.1016/j.eswa.2013.07.067 2-s2.0-84885950824 Sulaiman M. Salhi A. Selamoglu B. I. Kirikchi O. B. A plant propagation algorithm for constrained engineering optimisation problems Mathematical Problems in Engineering 2014 2014 10 627416 10.1155/2014/627416 2-s2.0-84901752748 Tellería J. L. Ramírez Á. Pérez-Tris J. Conservation of seed-dispersing migrant birds in Mediterranean habitats: shedding light on patterns to preserve processes Biological Conservation 2005 124 4 493 502 10.1016/j.biocon.2005.02.011 2-s2.0-17644379612 Wikipedia Contributors Strawberry 2015, http://bit.ly/17REoNP Ruth Strawberries eaten by animals, 2012, http://bit.ly/1zZg0jS Anisognathus Blue Winged Mountain-Tanager eating tree strawberries at San isidro Lodge 2012, http://bit.ly/1we3thV lifeisfull@ineedexchange.com Looking for nice strawberries to pick among the tiny ones, rotten ones and thistles 2014, https://lifeisfull.wordpress.com/page/5/ Kıran M. S. Gündüz M. A recombination-based hybridization of particle swarm optimization and artificial bee colony algorithm for continuous optimization problems Applied Soft Computing Journal 2013 13 4 2188 2203 10.1016/j.asoc.2012.12.007 2-s2.0-84886726325 Yao X. Liu Y. Lin G. Evolutionary programming made faster IEEE Transactions on Evolutionary Computation 1999 3 2 82 102 10.1109/4235.771163 2-s2.0-0032685734 Stork N. E. Lyal C. H. C. Extinction or ‘co-extinction’ rates? Nature 1993 366 6453 307 2-s2.0-0027378576 Jordano P. Fruits and frugivory Seeds: The Ecology of Regeneration in Plant Communities 2000 2 Wallingford, UK CABI 125 166 Debussche M. Isenmann P. Bird-dispersed seed rain and seedling establishment in patchy Mediterranean vegetation Oikos 1994 69 3 414 426 10.2307/3545854 2-s2.0-0028179029 Janzen D. H. Herbivores and the number of tree species in tropical forests The American Naturalist 1970 104 940 501 528 Levin S. A. Population dynamic models in heterogeneous environments Annual Review of Ecology and Systematics 1976 7 1 287 310 10.1146/annurev.es.07.110176.001443 Geritz S. A. H. de Jong T. J. Klinkhamer P. G. L. The efficacy of dispersal in relation to safe site area and seed production Oecologia 1984 62 2 219 221 10.1007/bf00379016 2-s2.0-0021558717 Levin S. A. Cohen D. Hastings A. Dispersal strategies in patchy environments Theoretical Population Biology 1984 26 2 165 191 10.1016/0040-5809(84)90028-5 MR770051 Augspurger C. K. Franson S. E. Wind dispersal of artificial fruits varying in mass, area, and morphology Ecology 1987 68 1 27 42 10.2307/1938802 2-s2.0-0023498834 Cooper R. B. Introduction to Queueing Theory 1972 Lawrence J. A. Pasternack B. A. Applied Management Science 2002 New York, NY, USA Wiley Ang A. H. S. Tang W. H. Probability concepts in engineering Planning 2004 1 4 1 3 Little J. D. C. A proof for the queuing formula: L=λW Operations Research 1961 9 3 383 387 10.1287/opre.9.3.383 Thompson D. W. On Growth and Form 1942 Courier van Houtan K. S. Pimm S. L. Halley J. M. Bierregaard R. O. Jr. Lovejoy T. E. Dispersal of Amazonian birds in continuous and fragmented forest Ecology Letters 2007 10 3 219 229 10.1111/j.1461-0248.2007.01004.x 2-s2.0-33847068710 Reynolds A. M. Frye M. A. Free-flight odor tracking in Drosophila is consistent with an optimal intermittent scale-free search PLoS ONE 2007 2 4, article e354 10.1371/journal.pone.0000354 2-s2.0-55849107781 Wikipedia Contributors Gamma function, 2015, http://bit.ly/1w6scza Sulaiman M. Salhi A. A seed-based plant propagation algorithm: the feeding station model Proceedings of the 5th International Conference on Metaheuristics and Nature Inspired Computing October 2014 Marrakesh, Morocco http://meta2014.sciencesconf.org/40158 Suganthan P. N. Hansen N. Liang J. J. Deb K. Chen Y. P. Auger A. Tiwari S. Problem definitions and evaluation criteria for the cec 2005 special session on real-parameter optimization 2005 Singapore Nanyang Technological University Liang J. J. Runarsson T. Mezura-Montes E. Clerc M. Suganthan P. N. Coello C. A. Deb K. Problem definitions and evaluation criteria for the CEC 2006, special session on constrained real-parameter optimization Journal of Applied Mechanics 2006 41 8 Karaboga D. Akay B. A modified artificial bee colony (abc) algorithm for constrained optimization problems Applied Soft Computing Journal 2011 11 3 3021 3031 10.1016/j.asoc.2010.12.001 2-s2.0-79951858405 He Q. Wang L. A hybrid particle swarm optimization with a feasibility-based rule for constrained optimization Applied Mathematics and Computation 2007 186 2 1407 1422 10.1016/j.amc.2006.07.134 MR2316936 2-s2.0-33947655944 Wang H.-Y. Ding X.-J. Cheng Q.-C. Chen F.-H. Lee M. Hirose A. Hou Z.-G. Kil R. M. An improved isomap for visualization and classification of multiple manifolds Neural Information Processing 2013 8227 Berlin, Germany Springer 1 12 Lecture Notes in Computer Science 10.1007/978-3-642-42042-9_1 Fogel D. B. System Identification Through Simulated Evolution: AMachine Learning Approach to Modeling 1991 Ginn Press Bäck T. Schwefel H.-P. An overview of evolutionary algorithms for parameter optimization Evolutionary Computation 1993 1 1 1 23 10.1162/evco.1993.1.1.1 Cagnina L. C. Esquivel S. C. Coello C. A. C. Solving engineering optimization problems with the simple constrained particle swarm optimizer Informatica 2008 32 3 319 326 2-s2.0-55749086218 Yang X.-S. Deb S. Engineering optimisation by cuckoo search International Journal of Mathematical Modelling and Numerical Optimisation 2010 1 4 330 343 10.1504/IJMMNO.2010.035430 2-s2.0-84858985277 Golinski J. An adaptive optimization system applied to machine synthesis Mechanism and Machine Theory 1973 8 4 419 436 10.1016/0094-114x(73)90018-9 2-s2.0-0015718527 Arora J. S. Introduction to Optimum Design 2004 Academic Press Belegundu A. D. Arora J. S. A study of mathematical programming methods for structural optimization. I. Theory International Journal for Numerical Methods in Engineering 1985 21 9 1583 1599 10.1002/nme.1620210904 MR810517 2-s2.0-0022130510