| Optimization Definition Properties | Chromosome definition | |
While Knowledge Builder allows you full control of all aspects of the genetic algorithm process, in practice, once you have defined your chromosomes you can run the optimization adequately with the default parameters setting. Modifying the evolution parameters is therefore a fine tuning exercise to improve the performance in two ways:
1.The speed of the optimization is often dictated by the number of chromosomes evaluated during a run. This in turn is a function of the number of generations, number of individuals per generation, the probability of adaptation and the number of adaptation cycles. Reducing any of these figures will speed up the optimization process and should therefore be considered if it does not significantly degrade the cost of the best solution.
2.If the optimization process does not converge on an effective solution then consider the following:
- increase the number of generations if the cost is steadily improving with advancing generations.
- increase the number of individuals per generation
- consider an adaptation strategy
A Multi-PC Optimization Engine is available that can use several PC's at once to work on very large optimization applications.
As an alternative to using Knowledge Builder @Commands for the main cost function (procedure), you can use VBScript syntax commands. These are recommended when large sequences of commands are needed in one procedure and the maximum execution speed at run time is required - typically the case with optimization applications.