Outline

 Introduction
 Artificial Bee Colony (ABC) Algorithm
 Foraging Behavior
 ABC Algorithm

1
Introduction
 Swarm Intelligence employs the collective
behaviors in the animal societies to design
algorithms.

 In 2005, Karaboga proposed an Artificial Bee

Colony (ABC), which is based on a particular
intelligent behavior of honeybee swarms.

2
Artificial Bee Colony (ABC)
 ABC is developed based on inspecting the
behaviors of real bees on finding nectar and
sharing the information of food sources to the
bees in the hive.

 Agents in ABC:
 The Employed Bee
 The Onlooker Bee
 The Scout

3
Artificial Bee Colony (ABC) (2)
 The Employed Bee:
It stays on a food source and provides the
neighborhood of the source in its memory.
 The Onlooker Bee:
It gets the information of food sources from
the employed bees in the hive and select one
of the food source to gathers the nectar.
 The Scout:
It is responsible for finding new food, the new
nectar, sources.
4
Artificial Bee Colony (ABC) (3)
 Procedures of ABC:
 Initialize (Move the scouts).
 Move the onlookers.
 Move the scouts only if the counters of the
employed bees hit the limit.
 Update the memory
 Check the terminational condition

5
Movement of the Onlookers
 Probability of Selecting a nectar source:
F  i 
Pi  S
(1)
 F  k 
k 1

Pi : The probability of selecting the ith employed

bee
S : The number of employed bees
θi : The position of the ith employed bee
F  i 
: The fitness value
6
Movement of the Onlookers (2)
 Calculation of the new position:
xij t  1   ij t     ij t    kj t  (2)
 xi : The position of the onlooker bee.
 t : The iteration number
  k : The randomly chosen employed bee.
 j : The dimension of the solution
   : A series of random variable in the range
. - 1, 1

7
Movement of the Scouts
 The movement of the scout bees follows
equation (3).
 ij   j min  r   j max   j min  (3)

 r : A random number and r  0,1

8
 Foraging Behavior

 Types of foraging bee

 Employed bees

 Unemployed bees
 Scout

 Onlooker bees

9
Picture form www.acclaimclipart.com
 Foraging Behavior (Cont.)

Hive

Dancing
area for A

Dancing
area for B

10

Picture form www.acclaimclipart.com , www.computerclipart.com

Behavior of Honey Bee Swarm

Three essential components of forage selection:

 Food Sources: The value of a food source depends

on many factors such as its proximity to the nest, its
richness or concentration of its energy, and the ease
of extracting this energy.

 Employed Foragers: They are associated with a

particular food source which they are currently
exploiting or are “employed” at. They carry with them
information about this particular source, its distance
and direction from the nest, the profitability of the
source and share this information with a certain
probability.
Behavior of Honey Bee Swarm

 Unemployed Foragers: They are continually at

look out for a food source to exploit. There are
two types of unemployed foragers: scouts,
searching the environment surrounding the nest
for new food sources and onlookers waiting in
the nest and establishing a food source through
the information shared by employed foragers.

 The model defines two leading modes of the

behavior:
 recruitment to a nectar source
 the abandonment of a source.
Exchange of Information among bees
 The exchange of information among bees is the
most important occurrence in the formation of
collective knowledge.

 The most important part of the hive with respect

to exchanging information is the dancing area

 Communication among bees related to the

quality of food sources takes place in the dancing
area.

 This dance is called a Waggle dance.

Exchange of Information among bees
 Employed foragers share their information with a
probability proportional to the profitability of the
food source, and the sharing of this information
through waggle dancing is longer in duration.

 An onlooker on the dance floor, probably she can

watch numerous dances and decides to employ
herself at the most profitable source.

 There is a greater probability of onlookers

choosing more profitable sources since more
information is circulated about the more
profitable sources.
15
16
Bees in Nature

17
18
Example
Basic Self Organization Properties
 Positive feedback: As the nectar amount of food
sources increases, the number of onlookers visiting
them increases, too.

 Negative feedback: The exploitation process of poor

food sources is stopped by bees.

 Fluctuations: The scouts carry out a random search

process for discovering new food sources.

 Multiple interactions: Bees share their information

about food sources with their nest mates on the
dance area.
Flowchart

21
ABC Algorithm
Explanation
 Each cycle of search consists of three steps: moving the
employed and onlooker bees onto the food sources and
calculating their nectar amounts; and determining the
scout bees and directing them onto possible food sources.

 A food source position represents a possible solution to

the problem to be optimized.

 The amount of nectar of a food source corresponds to the

quality of the solution

 Onlookers are placed on the food sources by using a

probability based selection process.
 Explanation
 As the nectar amount of a food source increases, the
probability value with which the food source is preferred by
onlookers increases, too.

 The scouts are characterized by low search costs and a low

average in food source quality. One bee is selected as the
scout bee.

 The selection is controlled by a control parameter called

"limit".

 If a solution representing a food source is not improved by a

predetermined number of trials, then that food source is
abandoned and the employed bee is converted to a scout.
ABC Algorithm
 ABC Algorithm

Foraging bee
employed bee
onlookers

Scout
 Initial Solutions

Food source initialize

(Number of solutions = Employed bees)
xi , j  xmin, j  rand (0,1)( xmax, j  xmin, j )

Where
i = 1,2,…,n
n = Food source
j = Dimension
 Employed bee Phase
 Send to each solutions (Can be done with Initial phase)
 Number of solutions = Employed bees

 Calculate fitness

f ( x1 )
f ( x2 )
f ( x3 )
f ( x4 )
f ( x5 )
 Employed bee Phase (Cont.)
 Evolve Solution to neighborhood Where
vij  xij  ij ( xij  xkj ) ij = rand(-1,1)
i = 1,2,…,n
Evolved
n = Food source
Solution Solution j = Dimension
Xi Vi k = rand(1,n)!=i
j=6
 Select better solution
 Onlookers Phase
 Calculate probability for each solution
f ( xi )
P{xi }  n

 f (x )
i 1
i

 Select solution due to probability

Employed bee 1 2 3 4 5

No i) ?
Ri<P(x

1 2 3 4 5 Ri =rand(0,1)
Onlooker
 Onlookers Phase (cont.)
 Evolve Solution to neighborhood
vij  xij  ij ( xij  xkj ) Where
ij= rand(-1,1)
i = 1,2,…,n
n = Food source
j = Dimension
k = rand(1,n)!=i
 Select better solution
(Same as Employed bee phase)
 Send scout

No of food source visited = “limit”

Send scouts to find new source

, else
 xmin  rand (0,1)( xmax  xmin ) , counter ≥ limit
xi (G  1)  
 xi (G )
Variations of ABC Algorithm
1. Basic ABC Algorithm:
 The basic ABC algorithm consists of three types of bees:
employed bees, onlooker bees, and scout bees.
2. Modified ABC:
 - Proposed to enhance the basic ABC algorithm's
performance.
- Different strategies for selecting employed and onlooker
bees, adaptive search strategies, or improved methods for
handling scout bees.
The goal is to increase the algorithm's efficiency and
convergence speed.
33
3. ABC with Local Search:
After generating a solution using the standard ABC mechanism, a
local search is applied to refine the solution.
Local search algorithms like hill climbing or gradient descent can
be used to explore the solution space around the current solution.
4. ABC with Different Neighborhoods:
 Instead of using a single neighborhood structure for all
employed bees, this variation of ABC employs different
neighborhoods or search strategies for different employed bees.
Each employed bee may focus on a specific region of the solution
space, potentially leading to a more diverse exploration and faster
convergence.
34
5. Adaptive ABC:
 Adaptive ABC algorithms dynamically adjust algorithm
parameters during the optimization process. Parameters such as the
number of scout bees, the size of neighborhoods, and the
exploration-exploitation balance are adapted based on the
algorithm's performance, leading to better adaptability to different
problem domains.
6. Hybrid ABC:
 Hybrid ABC algorithms combine the ABC algorithm with other
optimization techniques or metaheuristics to leverage their
strengths. For example, ABC can be combined with genetic
algorithms, particle swarm optimization, or simulated annealing to
create hybrid algorithms that aim to enhance the overall
optimization performance. 35
7. Multi-objective ABC:
 The standard ABC algorithm is designed for single-objective
optimization problems. Variations for multi-objective optimization
involve extending the ABC algorithm to handle multiple
conflicting objectives simultaneously. This often requires
additional mechanisms for selecting and maintaining a diverse set
of Pareto-optimal solutions.
8. Parallel ABC:
 In parallel ABC, multiple ABC instances or subpopulations run
concurrently. This can improve the exploration capability and
speed up the convergence process by enabling multiple search
threads to work on different parts of the solution space
simultaneously.
36
9. Memory-Based ABC:
 Memory-based ABC algorithms incorporate a memory
mechanism to store and reuse promising solutions. By maintaining
a repository of high-quality solutions found during the optimization
process, the algorithm can guide the search toward better regions of
the solution space.
10. Dynamic ABC:
 Dynamic ABC algorithms are designed to handle optimization
problems with changing landscapes or parameters. They adapt to
changes in the problem environment by adjusting their search
strategies or parameters in response to these changes.

37