Algorithm 1 Pseudo code of the dynsTGA metaheuristics

Initialization. Generate random initial population, set the iteration counter t to 1, the value of maximum iteration number ($MaxIter$), and the initial values of $\theta$ and $N_3$ control parameters

while $t<MaxGen$ do

Evaluate all solutions in the population and sort them according to their fitness value

for all solutions in $N_1$ do

Conduct local search by utilizing Equation (6)

Apply greedy selection mechanism to choose between the old and the new solution

end for

for all solutions in $N_2$ do

Move solutions towards the closest best solutions in $N_1$ by using Equations (7)–(9)

Apply greedy selection mechanism to choose between the old and the new solution

end for

Remove $N_3$ worst solutions from the population and replace them with randomly solutions from the search domain

Generate $N_4$ randomly distributed solutions and modify each solution in respect to the best solutions in $N_1$ by using the mask operator

Evaluate all solutions in the population and sort them according to their fitness value

Choose N best solutions as initial population for the next iteration

Recalculate the values of $\theta$ and $N_3$ along with $N_2$ and $N_1$ control parameters

end while

return The best solution from the population