Method for planning a semiconductor manufacturing process based on users' demands using a fuzzy system and a genetic algorithm model

1. A method for planning a semiconductor manufacturing process based on users' demands, comprising the steps of:

establishing a genetic algorithm model and inputting data, wherein a fitness function of the genetic algorithm model is formed by adding a plurality of cost functions, setting each cost function to have the same weight, defining the number of genetic evolution generations, and defining the number of chromosomes generated by each generation;

establishing a fuzzy system which has a plurality of input parameters and one output parameter, the output parameter representing percent difference of each cost function in neighbor generations;

setting to have a modulation parameter corresponding to each input parameter for adjusting fuzzy sets of the output parameter;

executing genetic algorithm actions;

executing fuzzing actions, fuzzy inference actions, and defuzzing actions;

eliminating the chromosomes that produces output parameter smaller than a defined lower limit, and defining the remaining chromosome that produce the largest output parameter as the optimum chromosome, wherein the genetic algorithm actions stops being executed upon the optimum chromosome; and

determining whether or not the number of the defined number of the generations has been reached, if yes, executing the optimum chromosome of the last generation; if no, continuing executing the genetic algorithm actions.

2. The method as claimed in claim 1 , wherein if there are more than two chromosomes that produce the largest output parameter, then further compare the values of the fitness function, wherein each chromosome that produces the fitness function with the smallest value is set as the optimum chromosome.

3. The method as claimed in claim 1 , wherein the larger the modulation parameters are set to be, the larger the adjusted ranges of the fuzzy sets of the output parameter are.

4. The method as claimed in claim 1 , wherein the fuzzy system is set to have three input parameters.

5. The method as claimed in claim 1 , wherein each input parameter is set to correspond to three fuzzy sets, and the output parameter is set to correspond to seven fuzzy sets.

6. The method as claimed in claim 1 , wherein a fuzzy rule of the fuzzy system is IF-THEN-type.

7. The method as claimed in claim 1 , wherein the defuzzing actions are executed via a centroid method.

8. The method as claimed in claim 1 , wherein the defuzzing actions are executed via a mean of maxima method.

9. The method as claimed in claim 1 , wherein the defuzzing actions are executed via a weighted average method.

10. A method for planning a semiconductor manufacturing process based on users' demands, which is applied in a lithography process, comprising the steps of:

establishing a genetic algorithm model and inputting data, wherein a fitness function of the genetic algorithm model is formed by adding a plurality of cost functions, setting each cost function to have the same weight, defining the number of genetic evolution generations, and defining the number of chromosomes generated by each generation;

establishing a fuzzy system which has a plurality of input parameters and one output parameter, the output parameter representing percent difference of each cost function in neighbor generations;

setting to have a modulation parameter corresponding to each input parameter for adjusting fuzzy sets of the output parameter;

executing genetic algorithm actions;

executing fuzzing actions, fuzzy inference actions and defuzzing actions;

eliminating the chromosomes that produces output parameter smaller than a defined lower limit, and defining the remaining chromosome that produce the largest output parameter as the optimum chromosome, wherein the genetic algorithm actions stops being executed upon the optimum chromosome; and

determining whether or not the number of the defined number of the generations has been reached, if yes, executing the optimum chromosomes of the last generation; if no, continuing executing the genetic algorithm actions.

11. The method as claimed in claim 10 , wherein if there are more than two chromosomes that produce the largest output parameter, then further compare the values of the fitness function, wherein each chromosome that produces the fitness function with the smallest value is set as the optimum chromosome.

12. The method as claimed in claim 10 , wherein the larger the modulation parameters are set to be, the larger the adjusted ranges of the fuzzy sets of the output parameter are.

13. The method as claimed in claim 10 , wherein the fuzzy system is set to have three input parameters.

14. The method as claimed in claim 10 , wherein each input parameter is set to correspond to three fuzzy sets, and the output parameter is set to correspond to seven fuzzy sets.

15. The method as claimed in claim 10 , wherein a fuzzy rule of the fuzzy system is IF-THEN-type.

16. The method as claimed in claim 10 , wherein the defuzzing actions are executed via a centroid method.

17. The method as claimed in claim 10 , wherein the defuzzing actions are executed via a mean of maxima method.

18. The method as claimed in claim 10 , wherein the defuzzing actions are executed via a weighted average method.