Page 64 - IJOCTA-15-1

O. Ayana, D. F. Kanbak, M. Kaya Keles / IJOCTA, Vol.15, No.1, pp.50-70 (2025)

Initialization: A population of sailfish and sar- x i old sail is the sailfish’s position at the current it-
dines is randomly created, and each sailfish and eration, α is a random number between 0 and 1
sardine are allocated a randomized position x i sail and λ i is a coefficient at the i th iteration that was
and y i in order where sail ∈ {sailfish}, sard produced as shown in Eq. 4:
sard
∈ {sardines} and i ∈ {number of iteration}. The λ i = 2 ∗ α ∗ PD − PD (4)
x i and y i positions suggest a possible solu-
sail sard
tion to the problem at i th iteration. where prey density, called PD, displays how many
preys there are during each iteration. Since there
is less prey when sailfish hunt in groups, the PD
Elitism Procedure: Each newly generated pop-
parameter is crucial for updating the sailfish’s po-
ulation is evaluated using the fitness function to
sition near to the prey school. PD parameter’s
determine the position of each agent search (sail-
formula is as shown in Eq. 5:
fish or sardine). When upgrading the positions of
N sail
search agents, useful solutions can often be lost, PD = 1 − (5)
and unless elitist selection is utilized, these posi- N sail + N sard
tions could not be as strong as the previous ones. where the number of sailfish and sardines in each
Elitism entails passing on the fittest solution(s) algorithm iteration is indicated by the variables
unchanged to the following generation. The sail- N sail and N sard .
fish’s best location is saved in each iteration of
the SOA algorithm and considered an elite. Dur- Furthermore, because the starting population of
ing the attack, the elite sailfish, which is the most sardines is usually bigger than that of sailfish,
physically fit sailfish to date, should be able to af- N sail is defined by N sard × P sard where P sard
fect the sardine’s acceleration and maneuverabil- is the sardine population’s percentage that forms
ity. As previously described, the slashing motion the starting population of sailfish. The λ param-
of the sailfish’s rostrum during group hunting will eter will be tended to 1 when α > 0.5, while it
also harm sardines. Because of this, the sailfish tends to -1 when α < 0.5 and it will be 0 if α=0.5.
will select the damaged sardine as their preferred The SOA algorithm shows any sailfish change its
target for collaborative hunting after each itera- position around the prey school in a 2-way man-
tion. The positions of the elite sailfish and injured ner. According to first method, when it comes
sardine with the highest fitness at the i th iter- to elite sailfish and injured sardines, sailfish have
ation, respectively, are x i elite sail and y i injured sard . the option of attacking the prey school. The sec-
Elitism procedure is designed to keep previously ond method involves the sailfish occupying empty
rejected solutions from being chosen again. space around the school of prey in order to resem-
ble surrounding the prey. In both cases, sailfish
Attack Strategy and Position Updating: Sailfish, hurt more sardines in the early stages of the hunt,
in fact, engage the prey when none of their peers which results in a higher success rate of capture
are attacking. In other words, the temporally co- in the collaborative hunting’s latter stages.
ordinated attack of sailfish can boost the num-
ber of hunting success. Sailfish herd and rush af- Hunting and Catching Prey and Position Updates
ter their prey. Sailfish herding behavior adjusts by the Sailfish: Both the sailfish power attack and
their position in relation to the location of other the sardine escape ability are usually fairly high
hunters around the prey school without cooper- at the start of the hunt. As a result, the sail-
ation. Sailfish adjust their position in a sphere fishes injure the sardines in school prey without
surrounding the optimal solution as a result. The being able to catch them in the early stages of
updated position of the sailfish in the SOA algo- the hunt. 79 The sailfishes’ assault power, as well
rithm during the i th iteration is shown in Eq. 3: as their ability to evade the sardines, declines
over time. Indeed, the sailfish blame the effort of
switching assault techniques, while the sardines
i i !
x elite sail + x injured sard accuse their bodies of damage. As a consequence,
x i new sail = x i elite sail − λ i α preys lose their capacity to avoid attacks in the
2 81
last stages of the hunt Hence, sailfish capture
− x i success rates are high. To consider sardine behav-
old sail
(3) ior when defending against sailfish attacks. The
following equation allows each sardine in SOA to
where x i is the position of elite sailfish,
elite sail update its position:
x i is the injured sardine’s best position, x i new sard = α ∗ (x i elite sail − x i old sard + AS) (6)
injured sard
58