Setting
|
Default
|
Range
|
Description
|
ND Points
(MAXNDP)
|
500
|
>0
|
The maximum number of non-dominated points which will be shown in the result file. If MAXNDP is less than the population size, then it will be reset to the number of population size.
|
Crowding Distance
(CRODIS)
|
0
|
0, 1, 2
|
Determines in which space the crowding distance is evaluated. Default value is 0 which means the crowding distance is evaluated in design space. 1 means the crowding distance is evaluated in solution space. 2 means the crowding distance is evaluated in both of the two spaces.
The crowding distance evaluation strategy allows users to get solutions more uniformly distributed in the selected space.
|
Discrete states
(GADISC)
|
1024
|
Integer > 1
|
Number of discrete values uniformly covering the range of continuous variables including upper and lower bound. Recommendation: select as a power of 2, e.g. 64 = 2^6, 1024 =2^10, etc.
A larger value allows for higher solution precision, but more computational effort is needed to find the optima.
|
Mutation rate
(GAMUTR)
|
0.01
|
0.0 – 1.0
|
Mutation rate (probability).
Larger value introduces more random effect. As a result, GA can explore more globally. But the convergence could be slower.
Recommended range: 0.001 - 0.05.
|
Elite population %
(GAELIT)
|
10
|
1.0 – 50.0
|
Percentage of population that belongs to elite.
The one with high fitness value is directly passed to the next generation. This is a very important strategy in GA to ensure the quality of solutions be non-decreasing. A larger value means that more individuals will be directly passed to the next generation. So a new gene has less of a chance to be introduced. The convergence speed could be increased. The drawback is that, too large a value could cause premature convergence.
Recommended range: 1.0 – 20.0.
|
Random Seed
(GAREPT)
|
0
|
Integer
0 to 10000
|
Controlling repeatability of runs depending on the way the sequence of random numbers is generated.
|
0
|
Random (non-repeatable).
|
> 0
|
Triggers a new sequence of pseudo-random numbers, repeatable if the same number is specified.
|
Number of contenders
(GATOUR)
|
2
|
Integer
2 to 5
|
Number of contenders in a tournament selection.
With a larger value, individuals with a lower fitness value have less of a chance of being selected. Thus, the good individuals have more chance to produce offspring. The bad effect is that diversity of the population is reduced. MOGA could converge prematurely.
|
Penalty power
(GAPPOW)
|
1
|
> 0.0,
< 10.0
|
Penalty power in the formulation of the fitness function as exterior penalty function.
Recommended range: 1.0 – 2.0.
|
Penalty multiplier
(GAPMUL)
|
2.0
|
> 0.0
|
Initial penalty multiplier in the formulation of the fitness function as exterior penalty function. Penalty multiplier will be increased gradually with iterating steps going on.
In general, a larger value allows the solution to become feasible with less iteration steps; but too large value could result in a worse solution.
Recommended range: 1.0 – 5.0.
|
Distribution Index
(GAINDX)
|
5
|
Integer
1 to 100
|
Distribution index used by real coded MOGA.
This parameter is to control offspring individuals to be close to or far away from the parent individuals. Increasing the value will result in offspring individuals being closer to the parents.
Recommended range: 3 – 10.
|
Type
(GATYPE)
|
0
|
0 or 1
|
0 Real coded MOGA is used.
1 Binary coded MOGA is used.
In general, real coded MOGA performs better than binary coded MOGA. For discrete optimization problem, binary coded MOGA could be better.
|
Hybrid Algorithm
|
No hybrid
|
No hybrid, Meta-model based method
|
No hybrid
|
Meta-model based method
|
Constraint threshold
(EPSCON)
|
1.0e-4
|
> 0.0
|
This parameter is used for constraint value calculation. In general, constraint value is normalized to its bound value. One exception is that, constraint value is not normalized if its absolute bound value is less than this parameter. Recommended range is 1.0e-6 ~ 1.0.
|
Use Inclusion Matrix
(INCLUSI)
|
No
|
No, With Initial
|
| • | No ignores the Inclusion matrix |
| • | With Initial combines Inclusion matrix with the initial population sample. |
|
