Controlling process modules using site models and monitor wafer control

Claim 1

A method of simultaneously controlling a plurality of nonuniformity measures in processing discrete products through a module consisting of a plurality individual processes, said method comprising: a. establishing a process model for a first process of said module by relating a plurality of product quality parameters to a plurality of process control variables to obtain an initial setting for said plurality of process control variables; b. determining goals for the output of a second process in said module subsequent to said first process which could be better achieved by changing the output of a previous process; c. changing the goals of said first process of said module to achieve the goals of said second process and establishing a set of process control variables for said first process; d. processing one or more products and measuring one or more product quality parameters on said products; e. exercising quality control tests on said products; f. if quality control tests do not indicate significant difference from said first process model prediction and goals then continuing processing at step g, otherwise continue by: (1) processing at least one monitor product to obtain observations for tuning using settings determined by local experiment design; (2) tuning said process models to create tuned process models using said plurality of product quality parameters from said products of step d, wherein said tuning estimates a changed state of said processing with the use of process models; (3) estimating new values for said plurality of process control variables from said tuned models; and (4) continuing processing products with said new values of said plurality of process control variables by repeating again at step d, substituting new products for said first plurality of products; g. continuing to process plurality of products through subsequent process of said process module; and h. monitoring the output of said second process and continuing processing a step g if said output of said second process is within process goals otherwise continuing with step b if said output of said second process is not within process goal tolerances.

Claim 2

A method of simultaneously controlling a plurality of nonuniformity measures in processing discrete products through a module consisting of a plurality individual processes, said method comprising: a. establishing a process model for a first process of said module by relating a plurality of product quality parameters to a plurality of process control variables to obtain an initial setting for said plurality of process control variables; b. determining goals for the output of a second process in said module subsequent to said first process which could be better achieved by changing the output of a previous process; c. changing the goals of said first process of said module to achieve the goals of said second process and establishing a set of process control variables for said first process; d. processing one or more products and measuring one or more product quality parameters on said products; e. exercising quality control tests on said products; f. if quality control tests do not indicate significant difference from said first process model prediction and goals then continuing processing at step g, otherwise continue by: (1) processing at least one monitor product to obtain observations for tuning using settings determined by local experiment design; (2) tuning said process models to create tuned process models using said plurality of product quality parameters from said products of step d, wherein said tuning estimates a changed state of said processing with the use of process models; (3) estimating new values for said plurality of process control variables from said tuned models; and (4) continuing processing products with said new values of said plurality of process control variables by repeating again at step d, substituting new products for said first plurality of products; g. continuing to process plurality of products through subsequent process of said process module; and h. monitoring the output of said second process and continuing processing a step g if said output of said second process is within process goals otherwise continuing with step b if said output of said second process is not within process goal tolerances.

Claim 12

A method of simultaneously controlling a plurality of nonuniformity measures in processing semiconductor wafers through a module consisting of a plurality individual processes, said method comprising: a. establishing a process model for a first process of said module by relating a plurality of product quality parameters to a plurality of process control variables to obtain an initial setting for said plurality of process control variables wherein said process models include site models and said site models model measures of nonuniformity and asymmetry as a functions of sites; b. determining goals for the output of a second process in said module subsequent to said first process which could be better achieved by changing the output of a previous process; c. changing the goals of said first process of said module to achieve the goals of said second process and establishing a set of process control variables for said first process; d. processing one or more wafers and measuring one or more product quality parameters on said products; e. exercising quality control tests on said wafers; f. if quality control tests do not indicate significant difference from said first process model prediction and goals then continuing processing at step g, otherwise continue by: (1) processing at least one monitor wafer to obtain observations for tuning using settings determined by local experiment design; (2) tuning said process models to create tuned process models using said plurality of product quality parameters from said wafers of step d, wherein said tuning estimates a changed state of said processing with the use of process models; (3) estimating new values for said plurality of process control variables from said tuned models; and (4) continuing processing wafers with said new values of said plurality of process control variables by repeating again at step d, substituting new wafers for said first plurality of products; g. continuing to process plurality of wafers through subsequent process of said process module; and h. monitoring the output of said second process and continuing processing a step g if said output of said second process is within process goals otherwise continuing with step b if said output of said second process is not within process goal tolerances.

Claim 13

A method of simultaneously controlling a plurality of nonuniformity measures in processing semiconductor wafers through a module consisting of a plurality individual processes, said method comprising: a. establishing a process model for a first process of said module by relating a plurality of product quality parameters to a plurality of process control variables to obtain an initial setting for said plurality of process control variables wherein said process models include site models and said site models model measures of nonuniformity and asymmetry as a functions of sites; b. determining goals for the output of a second process in said module subsequent to said first process which could be better achieved by changing the output of a previous process; c. changing the goals of said first process of said module to achieve the goals of said second process and establishing a set of process control variables for said first process; d. processing one or more wafers and measuring one or more product quality parameters on said products; e. exercising quality control tests on said wafers; f. if quality control tests do not indicate significant difference from said first process model prediction and goals then continuing processing at step g, otherwise continue by: (1) processing at least one monitor wafer to obtain observations for tuning using settings determined by local experiment design; (2) tuning said process models to create tuned process models using said plurality of product quality parameters from said wafers of step d, wherein said tuning estimates a changed state of said processing with the use of process models; (3) estimating new values for said plurality of process control variables from said tuned models; and (4) continuing processing wafers with said new values of said plurality of process control variables by repeating again at step d, substituting new wafers for said first plurality of products; g. continuing to process plurality of wafers through subsequent process of said process module; and h. monitoring the output of said second process and continuing processing a step g if said output of said second process is within process goals otherwise continuing with step b if said output of said second process is not within process goal tolerances.

Claim 19

A method of controlling a multi-step process module for manufacturing a product, said method comprising the steps: a. processing a raw material input with a controlled process with at least one controllable parameter to produce an intermediate product; b. processing said intermediate product with at least one uncontrolled process to produce a final product, at least one uncontrolled process producing nonuniform results across said final product; c. measuring characteristics of said final product to determine if the final product is within specification throughout said final product; d. adjusting at least one parameter of said controlled process to compensate for the nonuniform effect of step b if said final product is not in specification; and e. repeating steps a-d.

Claim 20

A method of controlling a multi-step process module for manufacturing a product, said method comprising the steps: a. processing a raw material input with a controlled process with at least one controllable parameter to produce an intermediate product; b. processing said intermediate product with at least one uncontrolled process to produce a final product, at least one uncontrolled process producing nonuniform results across said final product; c. measuring characteristics of said final product to determine if the final product is within specification throughout said final product; d. adjusting at least one parameter of said controlled process to compensate for the nonuniform effect of step b if said final product is not in specification; and e. repeating steps a-d.