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.

The present invention provides methods and systems for manufacturing process control of photovoltaic products. Some embodiments relate to a method for tracking wafers for photovoltaic products with respect to which production tool processed them and their position within that production tool. Some embodiments relate to measuring and characterizing the critical-to-quality parameters of the partially-finished photovoltaic products emerging from the production tool in question. Some embodiments relate to display and visualization of the measured parameters on a computer screen, such that the parameters of each production unit can be directly observed in the context of which production tools processed them, which location within a specific production tool they were located in during processing, and which batch, or in the case of continuous processing, what time, the unit(s) was/where processed.

An error diagnosis method includes: the parameter value obtaining step of obtaining multiple parameter values; the error detection step of calculating a Mahalanobis distance from a unit space based on the obtained parameter values and diagnosing whether or not error is caused at the real machine based on the calculated Mahalanobis distance; the error portion estimation step of estimating a error portion of the real machine based on the Mahalanobis distance calculated at the error detection step; and the matching determination step of structuring an error analyzing model for analyzing the real machine based on the error portion of the real machine estimated at the error portion estimation step and determining whether or not an output analytical signal of the real machine obtained by analysis of the error analyzing model and the output signal output from the real machine match with each other.

A computer-implemented method to control technical equipment that performs a production batch-run of a production process, the technical equipment providing data in a form of time-series from a set of data sources, the data sources being related to the technical equipment, includes: accessing a reference time-series with data from a previously performed batch-run of the production process, the reference time-series being related to a parameter for the technical equipment; and while the technical equipment performs the production batch-run: receiving a production time-series with data, identifying a sub-series of the reference time-series, and comparing the received time-series and the sub-series of the reference time-series, to provide an indication of similarity or non-similarity, in case of similarity, controlling the technical equipment during a continuation of the production batch-run, by using the parameter as control parameter.

A prognostic and health management system for precision ball grinding machines includes: a detector connected to a first and a second grinding discs; and a central processing module including a determining unit, a historical data recording unit, a timing unit, and an alarm unit. The detector is connected to the determining unit, detects data of the two grinding discs, and sends the data to the determining unit, the determining unit compares data in the historical data recording unit to determine whether the first and second grinding discs are abnormal, if there is an abnormality, the alarm unit releases a warning signal, if the determining unit determines that there is no abnormality, the timing unit records a grinding time of the first and second grinding discs, and based on a comparison of the past records, the alarm unit releases a warning signal after a predetermined time has elapsed.

Provided is a prediction apparatus including: a data acquisition unit configured to acquire setting value data indicating a setting value of a controlled object and physical quantity data indicating a physical quantity of a product obtained by controlling the controlled object; a prediction unit configured to calculate, using the setting value data and the physical quantity data, a plurality of prediction values obtained by predicting a plurality of physical quantities in the product on a basis of a setting value used for control of the controlled object; an evaluation unit configured to evaluate the plurality of prediction values on a basis of a predefined reference; and an output unit configured to output a setting value recommended according to a result of the evaluation.

A method includes receiving, from one or more sensors associated with manufacturing equipment, current trace data associated with producing, by the manufacturing equipment, a plurality of products. The method further includes performing signal processing to break down the current trace data into a plurality of sets of current component data mapped to corresponding component identifiers. The method further includes providing the plurality of sets of current component data and the corresponding component identifiers as input to a trained machine learning model. The method further includes obtaining, from the trained machine learning model, one or more outputs indicative of predictive data and causing, based on the predictive data, performance of one or more corrective actions associated with the manufacturing equipment.

Methods for controlling a converter for converting glass tubes to glass articles include preparing condition sets including settings for a plurality of process parameters, operating the converter to produce glass articles, measuring attributes of the glass articles, operating the converter at each of the condition sets, associating each glass article with a condition set used to produce the glass article and the attributes measured, developing operational models from the attributes measured and the condition sets, determining run settings for each of the plurality of process parameters based on the operational models, and operating the converter with each of the process parameters set to the run settings determined from the operational models.

A data processing apparatus includes a processor. The processor generates visualization data for displaying estimation results of manufacturing conditions based on estimation results and relationship data. The relationship data includes first relationship data as a relationship between first manufacturing conditions recorded during an analysis, and second relationship data as a relationship between second manufacturing conditions corresponding. The processor divides the estimation results of the manufacturing conditions into a first group based on the first relationship data, and into a second group based on the second relationship data. The processor generates the visualization data based on a change in manufacturing condition relationship between the first group and the second group.

To provide more test data during the manufacture of non-volatile memories and other integrated circuits, machine learning is used to generate virtual test values. Virtual test results are interpolated for one set of tests for devices on which the test is not performed based on correlations with other sets of tests. In one example, machine learning determines a correlation study between bad block values determined at die sort and photo-limited yield (PLY) values determined inline during processing. The correlation can be applied to interpolate virtual inline PLY data for all of the memory dies, allowing for more rapid feedback on the processing parameters for manufacturing the memory dies and making the manufacturing process more efficient and accurate. In another set of embodiments, the machine learning is used to extrapolate limited metrology (e.g., critical dimension) test data to all of the memory die through interpolated virtual metrology data values.