Systems and methods for controlling device performance variability during manufacturing of a device on wafers are disclosed. The system includes a process platform, on-board metrology (OBM) tools, and a first server that stores a machine-learning based process control model. The first server combines virtual metrology (VM) data and OBM data to predict a spatial distribution of one or more dimensions of interest on a wafer. The system further comprises an in-line metrology tool, such as SEM, to measure the one or more dimensions of interest on a subset of wafers sampled from each lot. A second server having a machine-learning engine receives from the first server the predicted spatial distribution of the one or more dimensions of interest based on VM and OBM, and also receives SEM metrology data, and updates the process control model periodically (e.g., to account for chamber-to-chamber variability) using machine learning techniques.

Systems, machines, and methods for monitoring wafer handling are disclosed herein. A system for monitoring wafer handling includes a sensor and a controller. The sensor is capable of being secured to an assembled wafer handling machine. The controller is in electronic communication with the sensor and includes control logic. The control logic is configured to store a reference output of the sensor when the wafer handling machine is aligned and is configured to generate an indication signal when a difference between the reference output and a current output of the sensor exceeds a threshold.

According to one embodiment, a controller calculates an estimation temperature profile, which represents a change with time in temperature of a substrate mounting surface in a state where a substrate is placed on a substrate stage, from warp amount information indicating a warp of the substrate. Further, the controller detects abnormality of a placement state of the substrate, on a basis of a difference between an actual measurement temperature profile, which represents a change with time in actual temperature measured by a temperature sensor in a state where the substrate is placed on the substrate stage, and the estimation temperature profile.

A lithographic process is performed on a plurality of semiconductor substrates. The method includes selecting one or more of the substrates as one or more sample substrates. Metrology steps are performed only on the selected one or more sample substrates. Based on metrology results of the selected one or more sample substrates, corrections are defined for use in controlling processing of the substrates or of future substrates. The selection of the one or more sample substrates is based at least partly on statistical analysis of object data measured in relation to the substrates. The same object data or other data can be used for grouping substrates into groups. Selecting of one or more sample substrates can include selecting substrates that are identified by the statistical analysis as most representative of the substrates in their group and/or include elimination of one or more substrates that are identified as unrepresentative.

An inspection system includes a controller communicatively coupled to a physical inspection device (PID), a virtual inspection device (VID) configured to analyze stored PID data, and a defect verification device (DVD). The controller may receive a pattern layout of a sample including multiple patterns fabricated with selected lithography configurations defining a process window, receive locations of PID-identified defects identified through analysis of the sample with the PID, wherein the PID-identified defects are verified by the DVD, remove one or more lithography configurations associated with the locations of the PID-identified defects from the process window, iteratively refine the process window by removing one or more lithography configurations associated with VID-identified defects identified through analysis of selected portions of stored PID data with the VID, and provide, as an output, the process window when a selected end condition is met.

There are provided a method of generating an inspection recipe usable for inspecting an inspection area of a specimen and a recipe generating unit. The recipe generating unit is configured: upon obtaining design data informative of design structural elements comprised in a design PoI corresponding to the at least one PoI, to provide global segmentation of a test image captured by an inspection tool unit from the inspection area and comprising at least one test PoI of substantially the same design as the at least one PoI, thereby to obtain segmented structural elements comprised in the test PoI and segmentation configuration data; to associate the segmented structural elements comprised in the test PoI with the design structural elements comprised in the design PoI, thereby to obtain design association data; and to generate an inspection recipe comprising, at least, segmentation configuration data and design association data.

A method includes receiving, into a measurement tool, a substrate having a material feature, wherein the material feature is formed on the substrate according to a design feature. The method further includes applying a source signal on the material feature, collecting a response signal from the material feature by using a detector in the measurement tool to obtain measurement data, and with a computer connected to the measurement tool, calculating a simulated response signal from the design feature. The method further includes, with the computer, in response to determining that a difference between the collected response signal and the simulated response signal exceeds a predetermined value, causing the measurement tool to re-measure the material feature.

Methods and metrology modules are provided, which derive landscape information (expressing relation(s) between metrology metric(s) and measurement parameters) from produced wafers, identifying therein indications for production process changes, and modify production process parameters with respect to the identified indications, to maintain the production process within specified requirements. Process changes may be detected in wafer(s), wafer lot(s) and batches, and the information may be used to detect root causes for the changes with respect to production tools and steps and to indicate tool aging and required maintenance. The information and its analysis may further be used to optimize the working point parameters, to optimizing designs of devices and/or targets and/or to train corresponding algorithms to perform the identifying, e.g., using training wafers.

Systems and methods for monitoring stability of a wafer inspection recipe over time are provided. One method includes collecting inspection results over time. The inspection results are generated by at least one wafer inspection tool while performing the wafer inspection recipe on wafers at different points in time. The method also includes identifying abnormal variation in the inspection results by comparing the inspection results generated at different times to each other. In addition, the method includes determining if the abnormal variation is attributable to the wafers, the wafer inspection recipe, or one or more of the at least one wafer inspection tool thereby determining if the wafer inspection recipe is stable over time.

A system for auto defect screening using adaptive machine learning includes an adaptive model controller, a defect/nuisance library and a module for executing data modeling analytics. The adaptive model controller has a feed-forward path for receiving a plurality of defect candidates in wafer inspection, and a feedback path for receiving defects of interest already screened by one or more existing defect screening models after wafer inspection. The adaptive model controller selects data samples from the received data, interfaces with scanning electron microscope (SEM) review/inspection to acquire corresponding SEM results that validate if each data sample is a real defect or nuisance, and compiles model training and validation data. The module of executing data modeling analytics is adaptively controlled by the adaptive model controller to generate and validate one or more updated defect screening models using the model training and validation data according to a target specification.