Algorithm 6: Tri-Criterion Selection Scheme
Inputs: $N$, $P=\left[X_1,\dots ,X_n,\dots ,X_N\right]$, $P^{off}=\left[X_1^{off},\dots ,X_n^{off},\dots .,X_N^{off}\right]$
01:	Initialize $P^{Next}\leftarrow \varnothing$;
02:	for $n=1$ to N do
03:	Assign $X_n.Impr=0$ for each n-th original learner stored in $P$;
04:	Calculate $X_n^{off}.Impr$ of every n-th offspring learner stored in $P^{off}$ with Equation (16);
05:	end for
06:	Construct the merged population $P^{MG}$ using Equation (17);
07:	Sort the solution members in $P^{MG}$ ascendingly based on fitness values;
08:	for $n=1$ to 2N do
09:	Calculate $X_n^{MG}.Dis$ of every n-th solution stored in $P^{MG}$ with Equation (18);
10:	end for
11:	Randomly generate the integers of $K_1\in \left\{1,N\right\}$, $K_2\in \left\{1,N-K_1\right\}$ and $K_3=N-K_1-K_2$;
12:	for $n=1$ to $K_1$ do /*Fitness criterion*/
13:	$X_n^{Next}\leftarrow X_n^{MG}$;
14:	$P^{Next}\leftarrow P^{Next}{\displaystyle \cup }{X}_n^{Next}$;
15:	end for
12:	for $n=K_1+1$ to 2N do
13:	Randomly generate $\alpha$ based on a normal distribution of $N\left(0.9,0.05\right)$;
14:	Restrict the value of $\alpha$ in between 0.8 and 1.
15:	Compute the $X_n^{MG}.WF$ of each n-th solution stored in $P^{MG}$ with Equation (19);
16:	Initialize the flag variable of each n-th solution stored in $P^{MG}$ as $X_n^{MG}.Flag=0$;
17:	end for
18:	for $n=K_1+1$ to $K_1+K_2$  do   /*Diversity criterion*/
19:	Randomly select $X_a^{MG}$ and $X_b^{MG}$ from $P^{MG}$, where $a,b\in \left\{K_1+1,2N\right\}$, $a\ne b$, and $X_a^{MG}.Flag=X_b^{MG}.Flag=0$.
20:	Determine $X_n^{Next}$ with Equation (20);
21:	$P^{Next}\leftarrow P^{Next}{\displaystyle \cup }{X}_n^{Next}$;
22:	if $X_a^{MG}$ is selected as $X_n^{Next}$ then  /*Prevent the selection of same solution members*/
23:	$X_a^{MG}.Flag=1$;
24:	else if $X_b^{MG}$ is selected as $X_n^{Next}$ then
25:	$X_b^{MG}.Flag=1$;
26:	end if
27:	end for
28:	for $n=K_1+K_2+1$ to $N$  do   /*Improvement rate criterion*/
29:	Randomly select $X_e^{MG}$ and $X_f^{MG}$ from $P^{MG}$, where $e,f\in \left\{K_1+1,2N\right\}$, e $\ne f$, and $X_e^{MG}.Flag=X_f^{MG}.Flag=0$.
30:	Determine $X_n^{Next}$ using Equation (21);
31:	$P^{Next}\leftarrow P^{Next}{\displaystyle \cup }{X}_n^{Next}$;
32:	if $X_e^{MG}$ is selected as $X_n^{Next}$ then /*Prevent the selection of same solution members*/
33:	$X_e^{MG}.Flag=1$;
34:	else if $X_f^{MG}$ is selected as $X_n^{Next}$ then
35:	$X_f^{MG}.Flag=1$;
36:	end if
37:	end for
Output:  $P^{Next}=\left[X_1^{Next},\dots ,X_n^{Next},\dots .,X_N^{Next}\right]$