A computing system may include a virtual cross metrology engine configured to construct a given virtual metrology model. The given virtual metrology model may take, as inputs, process parameters applied for the given step of a semiconductor fabrication process. The virtual cross metrology engine may also be configured to construct a subsequent virtual metrology model, and the subsequent step is performed after the given step in the semiconductor fabrication process. Doing so may include determining inputs for the subsequent virtual metrology model from a combination of the process parameters applied for the given step of the semiconductor fabrication process, process parameters applied for the subsequent step of the semiconductor fabrication process, and a wafer value for the given step of the semiconductor fabrication process that the given virtual metrology model is configured to predict.

The current disclosure describes techniques for managing vertical alignment or overlay in semiconductor manufacturing using machine learning. Alignments of interconnection features in a fan-out WLP process are evaluated and managed through the disclosed techniques. Big data and neural networks system are used to correlate the overlay error source factors with overlay metrology categories. The overlay error source factors include tool related overlay source factors, wafer or die related overlay source factors and processing context related overlay error source factors.

A method for exposing a pattern in an area on a surface using a charged particle beam lithography is disclosed and includes inputting an original set of exposure information for the area. The area comprises a plurality of pixels, and the original set of exposure information comprises dosages for the plurality of pixels in the area. A backscatter is calculated for a sub area of the area based on the original set of exposure information. A dosage for at least one pixel in a plurality of pixels in the sub area is increased, in a location where the backscatter of the sub area is below a pre-determined threshold, thereby increasing the backscatter of the sub area. A modified set of exposure information is output, including the increased dosage of the at least one pixel in the sub area.

A cooling controller receives, from one or more sensors, wafer information associated with a wafer. The cooling controller determines a pattern mask area for the wafer based on the wafer information. The cooling controller determines a cooling time for the wafer based on the pattern mask area. The cooling controller causes a cooling plate to cool the wafer for a time duration equal to the cooling time. Determining the cooling time for a wafer based on a pattern mask area provides stable and consistent wafer temperatures for wafers having different mask and layout properties, which reduces mask overlay variation and increases wafer yield.

A method for exposing a pattern in an area on a surface using a charged particle beam lithography is disclosed and includes inputting an original set of exposure information for the area. The area comprises a plurality of pixels, and the original set of exposure information comprises dosages for the plurality of pixels in the area. A backscatter is calculated for a sub area of the area based on the original set of exposure information including the dosages for the plurality of pixels in the area. An increase in dosage for at least one pixel in a plurality of pixels in the sub area is determined, in a location where the backscatter of the sub area is below a pre-determined threshold, thereby increasing the backscatter of the sub area.

A method for categorizing a substrate subject to a semiconductor manufacturing process including multiple operations, the method including: obtaining values of functional indicators derived from data generated during one or more of the multiple operations on the substrate, the functional indicators characterizing at least one operation; applying a decision model including one or more threshold values to the values of the functional indicators to obtain one or more categorical indicators; and assigning a category to the substrate based on the one or more categorical indicators.

A computing system may include a virtual cross metrology engine configured to construct a given virtual metrology model. The given virtual metrology model may take, as inputs, process parameters applied for the given step of a semiconductor fabrication process. The virtual cross metrology engine may also be configured to construct a subsequent virtual metrology model, and the subsequent step is performed after the given step in the semiconductor fabrication process. Doing so may include determining inputs for the subsequent virtual metrology model from a combination of the process parameters applied for the given step of the semiconductor fabrication process, process parameters applied for the subsequent step of the semiconductor fabrication process, and a wafer value for the given step of the semiconductor fabrication process that the given virtual metrology model is configured to predict.

An apparatus and method of diagnosing an unobserved operational parameter of a machine or apparatus. The method including obtaining a plurality of causal relationships between pairs of parameters of the machine or apparatus, wherein each pair includes a cause parameter and an effect parameter. For at least some of the parameters, a decomposition of the parameters into a plurality of information components is determined, based on the determined causal relationships between the parameters. The decomposition includes a synergistic information component including information obtained from a combination of at least two causal relationships having the parameter as effect parameter. A parameter is determined to include a negative synergistic information component. Based on the existence of the negative synergistic information component, it is diagnosed that an unobserved operational parameter provides a cause for the parameter including the negative synergistic information component.

In an example, a method includes operating, by a processor, on object model data. The object model data describes at least part of an object to be generated in additive manufacturing. The method also includes determining, by a processor, pattern data. The pattern data comprising areas of variable reflectance intended to be formed on a portion of the object. The method includes determining, by a processor, object generation instructions to apply a fusing agent to at least part of a layer of build material corresponding to the portion of the object in a density corresponding to the reflectance of the generated pattern data.

A control apparatus which generates control signal for controlling a control object, includes a first compensator configured to generate a first signal based on a control deviation of the control object, a corrector configured to generate a correction signal by correcting the control deviation in accordance with an arithmetic expression having an adjustable coefficient, a second compensator configured to generate a second signal by a neural network based on the correction signal, and an arithmetic device configured to generate the control signal based on the first signal and the second signal.