Multi-objective calibrations of lithography models

1. A method comprising:

by a computing system:

determining a candidate lithography model set from which to calibrate a lithography model according to multiple objectives, wherein the lithography model is configured to simulate a lithographic process for manufacturing an integrated circuit, wherein each of the multiple objectives is evaluated for the lithography model through a respective objective function, and wherein determining comprises:

initializing a population of parent candidate models, the parent candidate models derived based on the lithography model and each given parent candidate model having a respective model parameter value set for the lithography model;

generating child candidate models from the parent candidate models;

merging the parent candidate models and the child candidate models into a merged population of candidate models;

classifying the candidate models of the merged population into tiers of non-dominated fronts according to the respective objective functions for the multiple objectives;

determining a subset of the merged population based on the classified tiers of non-dominated fronts; and

identifying, as the candidate lithography model set, a Pareto-optimal front of the subset of the merged population determined based on the classified tiers; and

setting a given candidate lithography model in the candidate lithography model set as a calibrated lithography model for simulating the lithographic process.

2. The method of claim 1 , comprising performing multiple iterations of generating the child candidate models, merging the parent and child candidate models, classifying the candidate models of the merged population, and determining the subset of the merged population until a stopping criterion is satisfied,

wherein the subset of the merged population determined for a given iteration is used as the parent model population for a subsequent iteration.

3. The method of claim 1 , determining the subset of the merged population comprises sequentially adding candidate models into the subset on a tier by tier basis until a threshold number of candidate models has been added to the subset.

4. The method of claim 3 , wherein determining the subset of the merged population further comprises selecting some, but not all, of the candidate models of a given tier of non-dominated fronts to add to the subset based on a computed distance between the candidate models and neighboring candidate models of the given tier.

5. The method of claim 1 , wherein the multiple objectives comprise a primary objective evaluated through a root mean square-based objective function and a secondary objective evaluated through a maximal critical dimension error-based objective function.

6. The method of claim 1 , wherein the multiple objectives comprise a primary objective evaluated for a full set of gauges of physical measurement data used to evaluate the lithography model a secondary objective evaluated for a selected subset of the gauges.

7. The method of claim 1 , further comprising:

providing a visual comparison of candidate lithography models in the candidate lithography model set evaluated for the multiple objectives; and

determining the given candidate lithography model to set as the calibrated lithography model based on user input received for the visual comparison.

8. A system comprising:

a model calibration engine configured to determine a candidate lithography model set from which to calibrate a lithography model according to multiple objectives, wherein the lithography model is configured to simulate a lithographic process for manufacturing an integrated circuit, wherein each of the multiple objectives is evaluated for the lithography model through a respective objective function, and wherein the model calibration engine is configured to determine the candidate lithography model set by:

initializing a population of parent candidate models, the parent candidate models derived based on the lithography model and each given parent candidate model having a respective model parameter value set for the lithography model;

generating child candidate models from the parent candidate models;

merging the parent candidate models and the child candidate models into a merged population of candidate models;

classifying the candidate models of the merged population into tiers of non-dominated fronts according to the respective objective functions for the multiple objectives;

determining a subset of the merged population based on the classified tiers of non-dominated fronts; and

identifying, as the candidate lithography model set, a Pareto-optimal front of the subset of the merged population determined based on the classified tiers; and

a model selection engine configured to set a given candidate lithography model in the candidate lithography model set as a calibrated lithography model for simulating the lithographic process.

9. The system of claim 8 , wherein the model calibration engine is configured to perform multiple iterations of generating the child candidate models, merging the parent and child candidate models, classifying the candidate models of the merged population, and determining the subset of the merged population until a stopping criterion is satisfied,

wherein the subset of the merged population determined for a given iteration is used as the parent model population for a subsequent iteration.

10. The system of claim 8 , wherein the model calibration engine is configured to determine the subset of the merged population by sequentially adding candidate models into the subset on a tier by tier basis until a threshold number of candidate models has been added to the subset.

11. The system of claim 10 , wherein the model calibration engine is configured to determine the subset of the merged population further by selecting some, but not all, of the candidate models of a given tier of non-dominated fronts to add to the subset based on a computed distance between the candidate models and neighboring candidate models of the given tier.

12. The system of claim 8 , wherein the multiple objectives comprise a primary objective evaluated through a root mean square-based objective function and a secondary objective evaluated through a maximal critical dimension error-based objective function.

13. The system of claim 8 , wherein the multiple objectives comprise a primary objective evaluated for a full set of gauges of physical measurement data used to evaluate the lithography model a secondary objective evaluated for a selected subset of the gauges.

14. The system of claim 8 , wherein the model selection engine is further configured to:

cluster the candidate lithography models in the candidate lithography model set based on model similarity; and

determine the given candidate lithography model to set as the calibrated lithography model based on the clustering.

15. A non-transitory machine-readable medium comprising instructions that, when executed by a processor, cause a computing system to:

determine a candidate lithography model set from which to calibrate a lithography model according to multiple objectives, wherein the lithography model is configured to simulate a lithographic process for manufacturing an integrated circuit, wherein each of the multiple objectives is evaluated for the lithography model through a respective objective function, and wherein determination of the candidate lithography model set comprises:

initializing a population of parent candidate models, the parent candidate models derived based on the lithography model and each given parent candidate model having a respective model parameter value set for the lithography model;

generating child candidate models from the parent candidate models;

merging the parent candidate models and the child candidate models into a merged population of candidate models;

classifying the candidate models of the merged population into tiers of non-dominated fronts according to the respective objective functions for the multiple objectives;

determining a subset of the merged population based on the classified tiers of non-dominated fronts; and

identifying, as the candidate lithography model set, a Pareto-optimal front of the subset of the merged population determined based on the classified tiers; and

set a given candidate lithography model in the candidate lithography model set as a calibrated lithography model for simulating the lithographic process.

16. The non-transitory machine-readable medium of claim 15 , wherein the instructions further cause the computing system to perform multiple iterations of generating the child candidate models, merging the parent and child candidate models, classifying the candidate models of the merged population, and determining the subset of the merged population until a stopping criterion is satisfied,

wherein the subset of the merged population determined for a given iteration is used as the parent model population for a subsequent iteration.

17. The non-transitory machine-readable medium of claim 15 , wherein the instructions cause the computing system to determine the subset of the merged population by sequentially adding candidate models into the subset on a tier by tier basis until a threshold number of candidate models has been added to the subset.

18. The non-transitory machine-readable medium of claim 17 , wherein the instructions cause the computing system to determine the subset of the merged population further by selecting some, but not all, of the candidate models of a given tier of non-dominated fronts to add to the subset based on a computed distance between the candidate models and neighboring candidate models of the given tier.

19. The non-transitory machine-readable medium of claim 15 , the multiple objectives comprise:

a primary objective evaluated through a root mean square-based objective function and a secondary objective evaluated through a maximal critical dimension error-based objective function; or

a primary objective evaluated for a full set of gauges of physical measurement data used to evaluate the lithography model a secondary objective evaluated for a selected subset of the gauges.

20. The non-transitory machine-readable medium of claim 15 , wherein the instructions further cause the computing system to:

provide a visual comparison of candidate lithography models in the candidate lithography model set evaluated for the multiple objectives;

cluster the candidate lithography models in the candidate lithography model set based on model similarity; and

determine the given candidate lithography model to set as the calibrated lithography model based on user input received for the visual comparison or the clustering.