Some implementations of the disclosure are directed to a method, comprising: receiving tapped data that was tapped from a controller of an industrial control system (ICS) while the controller executed first control code to control ICS devices of the ICS, the tapped data used during execution of the first control code, and the tapped data comprising input data obtained from one or more input components of the controller communicatively coupled to the ICS devices, or output data obtained from one or more output components of the controller communicatively coupled to the ICS devices; and after receiving the tapped data, executing, using at least the tapped data, second control code that provides an emulation of the controller, the emulation comprising running, using at least the tapped data, a process of the first control code at a faster rate than it is run by the controller executing the first control code.

A diagnosis system diagnoses presence or absence of an abnormality from data pieces collected in a factory. The diagnosis system includes (i) a diagnoser that diagnoses presence or absence of an abnormality by classifying, in accordance with a diagnosis model defining a plurality of groups, the collected data pieces into at least one of the plurality of groups, (ii) an extractor that extracts, from the collected data pieces, a candidate for a data piece to belong to a new group different from the plurality of groups, (iii) a reception device that provides candidate information relating to the candidate extracted by the extractor, (iv) and a learner that learns a new model including the new group. The diagnoser diagnoses presence or absence of an abnormality with the new model after the new model is learned.

A method of the present disclosure includes (a) retrieving from a memory a plurality of measured values of the factor variable, and a label indicating good or bad of the quality corresponding to each of the plurality of measured values, (b) dividing a factor variable space defined by the factor variable into a plurality of grids by equally dividing a range determined by a maximum value and a minimum value of the plurality of measured values for each factor variable, (c) setting a plurality of candidate areas each of which includes one grid or a plurality of adjacent grids, and deriving, for each of the plurality of candidate areas, a good density based on the label associated with the measured value that is within the candidate area, and (d) selecting one of the plurality of candidate areas as the factor variable area, based on the good density.

Embodiments disclosed herein comprise a sensor. In an embodiment, the sensor comprises a substrate having a first surface and a second surface opposite from the first surface. In an embodiment, the sensor further comprises a first electrode over the first surface of the substrate, and a second electrode over the first surface of the substrate and adjacent to the first electrode. In an embodiment, the sensor further comprises a barrier layer over the first electrode and the second electrode.

An information processing apparatus includes an acquisition unit configured to acquire process information about a substrate process, the process information including process data and a process condition, and a display control unit configured to control a display on a display apparatus based on the process information acquired by the acquisition unit, wherein the display control unit selectively displays, on the display apparatus, a first screen that displays the process data of a lot including a plurality of substrates on a lot-by-lot basis and a second screen that displays the process data of a first lot on a substrate-by-substrate basis, the first lot being a lot designated by a user from the lot displayed on the first screen.

Disclosed is a method of determining at least one tolerance band limit value for a technical variable under test. The method includes obtaining the at least one tolerance band limit value from sample tolerance band limit values of different samples, wherein the samples comprise values of the technical variable under test of the associated sample, wherein obtaining the at least one tolerance band limit value comprises using a location measure of a distribution according to which the sample tolerance band limit values are distributed, wherein the technical variable under test is distributed according to an underlying extreme value distribution function, wherein each of the sample tolerance band limit values is calculable using a sample-specific conditional probability distribution function which is a function of sample values of the sample, and wherein the technical variable relates to a physical characteristic of a product that is producible in an industrial mass production process.

A method of detecting and classifying anomalies during semiconductor processing includes executing a wafer recipe a semiconductor processing system to process a semiconductor wafer; monitoring sensor outputs from a sensors that monitor conditions associated with the semiconductor processing system; providing the sensor outputs to models trained to identify when the conditions associated with the semiconductor processing system indicate a fault in the semiconductor wafer; receiving an indication of a fault from at least one of the models; and generating a fault output in response to receiving the indication of the fault.

Industrial quality control is challenging for artificial neural networks (ANNs) and deep neural networks (DNNs) because of the nature of the processed data: there is an abundance of consistent data representing good products, but little data representing bad products. In quality control, the task is changed from conventional DNN task of “recognize what I learned best” to “recognize what I have never seen before.” Lifelong DNN (L-DNN) technology is a hybrid semi-supervised neural architecture that combines the ability of DNNs to be trained, with high precision, on known classes, while being sensitive to any number of unknown classes or class variations. When used for industrial inspection, L-DNN exploits its ability to learn with little and highly unbalanced data. L-DNN's real-time learning capability takes advantage of rare cases of poor-quality products that L-DNN encounters after deployment. L-DNN can be applied to industrial inspections and manufacturing quality control.

Systems and methods for data collection for an industrial heating process are disclosed. The system according to one embodiment can include a plurality of data collectors, including a swarm of self-organized data collector members, wherein the swarm of self-organized data collector members organize to enhance data collection based on at least one of capabilities and conditions of the data collector members of the swarm, and wherein the plurality of data collectors is coupled to a plurality of input channels for acquiring collected data relating to the industrial heating process, and a data acquisition and analysis circuit for receiving the collected data via the plurality of input channels and structured to analyze the received collected data using a neural network to monitor a plurality of conditions relating to the industrial heating process.

Methods, systems, and non-transitory computer readable medium are described for sensor metrology data integration. A method includes receiving sets of sensor data and sets of metrology data. Each set of sensor data includes corresponding sensor values associated with producing corresponding product by manufacturing equipment and a corresponding sensor data identifier. Each set of metrology data includes corresponding metrology values associated with the corresponding product manufactured by the manufacturing equipment and a corresponding metrology data identifier. The method further includes determining common portions between each corresponding sensor data identifier and each corresponding metrology data identifier. The method further includes, for each of the sensor-metrology matches, generating a corresponding set of aggregated sensor-metrology data and storing the sets of aggregated sensor-metrology data to train a machine learning model. The trained machine learning model is capable of generating one or more outputs for performing a corrective action associated with the manufacturing equipment.