System and method for learning and/or optimizing manufacturing processes
A system and method for learning and/or optimizing processes related to semiconductor manufacturing is provided. A learning component generates a set of candidate process models based on process data associated with one or more fabrication tools. The learning component also selects a particular process model from the set of candidate process models that is associated with lowest error. An optimization component generates a set of candidate solutions associated with the particular process model. The optimization component also selects a particular solution from the set of candidate solutions based on a target output value and an output value associated with the particular solution.
1. A system, comprising:
a memory storing computer executable components; and
a processor configured to execute the following computer executable components stored in the memory:
a learning component that generates a set of candidate process models based on process data associated with one or more fabrication tools and selects a particular process model from the set of candidate process models that is associated with lowest error; and
an optimization component that generates a set of candidate solutions associated with the particular process model and selects a particular solution from the set of candidate solutions based on a target output value and an output value associated with the particular solution.
2. The system of claim 1 , wherein the process data is associated with a data matrix that includes at least one of sensor measurement data, spectral data, device measurement data and maintenance data.
3. The system of claim 1 , wherein the learning component generates a quality value for each candidate process model in the set of candidate process models to facilitate selection of the particular process model from the set of candidate process models.
4. The system of claim 3 , wherein the quality value for each candidate process model in the set of candidate process models is associated with mean squared error.
5. The system of claim 3 , wherein the learning component further generates a diversity value associated with uniqueness for each candidate process model in the set of candidate process models.
6. The system of claim 5 , wherein the diversity value for each candidate process model in the set of candidate process models is generated based on a cosine function.
7. The system of claim 5 , wherein the learning component generates a ranking of the set of candidate process models as a function of the quality value and the diversity value for each candidate process model in the set of candidate process models.
8. The system of claim 7 , wherein the learning component further selects one or more other process models from the set of candidate process models based on the ranking, and stores the particular process model and the one or more other process models in a short term model memory.
9. The system of claim 1 , wherein the optimization component generates a quality value for each candidate solution in the set of candidate solutions to facilitate selection of the particular solution from the set of candidate solutions.
10. The system of claim 9 , wherein the quality value for each candidate solution in the set of candidate solutions is associated with mean squared error.
11. The system of claim 9 , wherein the optimization component further generates a diversity value associated with uniqueness for each candidate solution in the set of candidate solutions.
12. The system of claim 11 , wherein the diversity value for each candidate solution in the set of candidate solutions is generated based on a cosine function.
13. The system of claim 11 , wherein the optimization component generates a ranking of the set of candidate solutions as a function of the quality value and the diversity value for each candidate solution in the set of candidate solutions.
14. The system of claim 13 , wherein the optimization component further selects one or more other solutions from the set of candidate solutions based on the ranking, and stores the particular solution and the one or more other solutions in a short term optimization memory.
15. A method, comprising:
employing a processor that facilitates execution of computer executable instructions stored on a non-transitory computer readable medium to implement operations, comprising:
generating a set of candidate process models based on process data associated with one or more fabrication tools;
generating a quality value for each candidate process model in the set of candidate process models;
selecting a particular process model from the set of candidate process models based on the quality value associated with each candidate process model in the set of candidate process models;
generating a set of candidate solutions associated with the particular process model; and
selecting a particular solution from the set of candidate solutions based on a target output value and an output value associated with the particular solution.
16. The method of claim 15 , further comprising:
generating a diversity value for each candidate process model in the set of candidate process models.
17. The method of claim 16 , further comprising:
ranking the set of candidate process models based on the quality value and the diversity value for each candidate process model in the set of candidate process models;
selecting a subset of candidate process models based on the ranking of the set of candidate process models; and
storing the subset of candidate process models in a short term model memory.
18. The method of claim 15 , further comprising:
generating a quality value and a diversity value for each candidate solution in the set of candidate solutions.
19. The method of claim 18 , further comprising:
ranking the set of candidate solutions based on the quality value and the diversity value for each candidate solution in the set of candidate solutions;
selecting a subset of candidate solutions based on the ranking of the set of candidate solutions; and
storing the subset of candidate solutions in a short term optimization memory.
20. A computer-readable medium having stored thereon computer-executable instructions that, in response to execution by a system including a processor, cause the system to perform operations, the operations including:
generating a set of candidate process models based on process data associated with one or more fabrication tools;
generating a quality value for each candidate process model in the set of candidate process models;
selecting a particular process model from the set of candidate process models based on the quality value associated with each candidate process model in the set of candidate process models;
generating a set of candidate solutions associated with the particular process model; and
selecting a particular solution from the set of candidate solutions based on a target output value and an output value associated with the particular solution.