Advanced-process-control system utilizing a lambda tuner

1. An advanced-process-control system, comprising:

a database for receiving process data from a measurement tool for a plurality of process runs and for storing said process data;

a lambda tuner for determining a tuned-lambda value corresponding to a process-capability-index value based on upper and lower process control limits and statistics derived from said process data; and

a process-run controller for updating a recipe value based on said process data and said tuned-lambda value;

wherein said lambda tuner further comprises:

a run-to-run simulator for simulating process-equipment-output values, said run-to-run simulator applied to said process data, a process-equipment-output model, and process-equipment-model coefficients for simulating said process-equipment-output values, and for providing to said lambda tuner a plurality of lambda values and corresponding process-equipment-output values for a plurality of simulated process runs;

wherein said lambda tuner further comprises:

a tuned-lambda selector for selecting said tuned-lambda value, said tuned-lambda selector comprising:

a process-capability-index calculator for calculating said statistics from said process-equipment-output values, and for calculating said process-capability-index value from said statistics as a function of said lambda value; and

a moving-average, process-capability-index calculator for calculating moving-average, process-capability-index values from said process-capability-index values to smooth out variations in said process-capability-index values, said tuned-lambda selector for selecting as said tuned-lambda value a selected lambda value corresponding to a selected moving-average, process-capability-index value in a plurality of said moving-average, process-capability-index values.

2. The advanced-process-control system according to claim 1 , further comprising:

a strategy controller for sending strategy-control information to said lambda tuner and said process-run controller, said strategy-control information comprising a process-equipment-model selection command and a simulation-mode selection command.

3. The advanced-process-control system according to claim 2 , wherein said strategy controller further comprises a human-machine interface to said advanced-process-control system for receiving instructions to send said strategy-control information to said lambda tuner and said process-run controller and for communicating information sent from said lambda tuner and said process-run controller.

4. The advanced-process-control system according to claim 1 , wherein said lambda tuner further comprises:

a tuned-lambda selector for selecting said tuned-lambda value, said tuned-lambda selector comprising:

a process-capability-index calculator for calculating said statistics from said process-equipment-output values, and for calculating said process-capability-index value from said statistics as a function of said lambda value, said tuned-lambda selector for selecting as said tuned-lambda value a selected lambda value corresponding to a selected process-capability-index value in a plurality of said process-capability-index values.

5. The advanced-process-control system according to claim 1 , wherein said lambda tuner further comprises:

a tuned-lambda selector for selecting said tuned-lambda value, said tuned-lambda selector comprising:

a process-capability-index calculator for calculating said statistics from said process-equipment-output values, and for calculating said process-capability-index values from said statistics as a function of said lambda values;

wherein said statistics comprise process-equipment-output, mean values, μ yi , and process-equipment-output, standard-deviation values, σ yi ; and

wherein said process-capability-index values, C pk i , are given by:

C pk i =min{ UCL−μ yi , μ yi −LCL}/ 3σ yi

wherein UCL is an upper control limit, and LCL is a lower control limit for process-equipment-output values.

6. The advanced-process-control system according to claim 1 , further comprising:

a process-run selector accessible to said lambda tuner and said process-run controller for selecting a plurality of process runs spanning a run-time window, said run-time window comprising an adjustable run-time-window width spanning a range of successive run times.

7. The advanced-process-control system according to claim 1 , further comprising:

a run-time-window selector accessible to said lambda tuner and said process-run controller for selecting a run-time window from a plurality of run-time windows, said run-time window in said plurality of run-time windows separated from an adjacent run-time window in said plurality of run-time windows by an inter-window interval, said inter-window interval comprising an adjustable inter-window-interval width spanning a range of successive run times between said run-time window and said adjacent run-time window.

8. The advanced-process-control system according to claim 1 , wherein said lambda tuner further comprises:

a process-data synthesizer for synthesizing synthetic process data, said synthetic process data comprising a waveform as a function of run time selected from the group of waveforms consisting of: a ramp, independently and identically distributed data, brown noise, a periodically repeating step, a sine wave, a periodically repeating impulse, a sawtooth, and a constant level.

9. A method for advanced process control, comprising:

storing process data received from a measurement tool in a database;

calculating process-capability-index values based on upper and lower process control limits and statistics derived from said process data;

determining a tuned-lambda value corresponding to a value in said process-capability-index values; and

calculating an updated recipe value based on said process data and said tuned-lambda value;

further comprising:

executing a run-to-run-simulation applied to said process data, a process-equipment-output model, and process-equipment-model coefficients for simulating process-equipment-output values as a function of lambda values;

further comprising:

calculating said statistics from said process-equipment-output values and calculating said process-capability-index values from said statistics as a function of said lambda values;

calculating moving-average, process-capability-index values from said process-capability-index values as a function of said lambda values to smooth out variations in said process-capability-index values; and

selecting as said tuned-lambda value a selected lambda value corresponding to a selected moving-average, process-capability-index value in a plurality of said moving-average, process-capability-index values.

10. The method for advanced process control according to claim 9 , further comprising:

sending strategy-control information to a lambda tuner, said strategy-control information comprising a process-equipment-model selection command and a simulation-mode selection command;

sending strategy-control information to a process-run controller, said strategy-control information comprising said process-equipment-model selection command; and

sending information to a strategy controller, said information comprising run-to-run simulation data and process-data, control-chart information.

11. The method for advanced process control according to claim 9 , further comprising:

calculating said statistics from said process-equipment-output values and calculating said process-capability-index values from said statistics as a function of said lambda values; and

selecting as said tuned-lambda value a selected lambda value corresponding to a selected process-capability-index value in a plurality of said process-capability-index values.

12. The method for advanced process control according to claim 9 , further comprising:

calculating said statistics from said process-equipment-output values, and calculating said process-capability-index values from said statistics as a function of said lambda values;

wherein said statistics comprise process-equipment-output, mean values, μ yi , and process-equipment-output, standard-deviation values, σ yi ; and

wherein said process-capability-index values, C pk i , are given by:

C pk i =min{ UCL−μ yi , μ yi −LCL}/ 3σ yi

wherein UCL is an upper control limit, and LCL is a lower control limit for process-equipment-output values.

13. The method for advanced process control according to claim 9 , further comprising:

adjusting a run-time-window width of a run-time window spanning a range of run times of process runs; and

selecting a plurality of said process runs spanned by said run-time window.

14. The method for advanced process control according to claim 9 , further comprising:

adjusting an inter-window-interval width of an inter-window interval spanning a range of successive run times between a run-time window and an adjacent run-time window, wherein said run-time window and said adjacent run-time window are permitted to be superimposed; and

selecting said run-time window from a plurality of run-time windows separated by a plurality of inter-window intervals.

15. The method for advanced process control according to claim 9 , further comprising:

synthesizing synthetic process data comprising a waveform as a function of run time selected from the group of waveforms consisting of: a ramp, independently and identically distributed data, brown noise, a periodically repeating step, a sine wave, a periodically repeating impulse, a sawtooth, and a constant level.

16. An article of manufacture comprising a program storage medium readable by a computer, the medium tangibly embodying a program of instructions executable by the computer to perform a method for advanced process control, comprising:

storing process data received from a measurement tool in a database;

calculating process-capability-index values based on upper and lower process control limits and statistics derived from said process data;

determining a tuned-lambda value corresponding to a value in said process-capability-index values; and

calculating an updated recipe value based on said process data and said tuned-lambda value;

further comprising:

executing a run-to-run-simulation applied to said process data, a process-equipment-output model, and process-equipment-model coefficients for simulating process-equipment-output values as a function of lambda values;

further comprising:

calculating said statistics from said process-equipment-output values and calculating said process-capability-index values from said statistics as a function of said lambda values;

calculating moving-average, process-capability-index values from said process-capability-index values as a function of said lambda values to smooth out variations in said process-capability-index values; and

selecting as said tuned-lambda value a selected lambda value corresponding to a selected moving-average, process-capability-index value in a plurality of said moving-average, process-capability-index values.