Abstract: A method includes obtaining training data from an operation of an industrial process plant; using the training data to train a machine learning model for assisting a plant operation of the industrial process plant; determining several sub-populations in the training data, each relating to a different scenario of an operation of the industrial process plant; determining several performances of the machine learning model, each relating to a different sub-population; determining, based on the performance of each one of the sub-populations, one or more underrepresented sub-populations; and revising, based on the one or more underrepresented sub-populations, the machine learning model for increased consideration of the one or more underrepresented sub-populations when assisting the plant operation of the industrial process plant.

Abstract: A computer-implemented method for assessing trustability of operator feedback from an operator of a control system of an industrial plant, comprising obtaining current error data from an error prediction model of the control system, wherein the current error data is indicative of a current error in the industrial plant, and wherein the error prediction model is based on machine learning; obtaining different error data that is indicative of one or more different errors, which are different from the current error; providing an operator request on an operator interface of the control system, wherein the operator request relates to the current error and the one or more different errors; obtaining operator feedback from the operator on the operator interface in response to the operator request provided thereon; and assessing the trustability of the operator feedback based on the operator feedback, the current error data and the different error data.

Abstract: A method for automatically generating a textual description of time series for monitoring or forecasting a process variable, comprising training a deep learning model on a cross-modal autoencoding module having an architecture consisting of a time series encoder, a text decoder and a time series decoder, obtaining time series data by invoking a display of readings of the temporal process variable, encoding, using the time series encoder and based on the deep learning model, the time series data and generating an embedding as input for the text decoder, transferring the generated embedding from the time series encoder to the text decoder, and generating, using the text decoder and based on the deep learning model, the textual description of time series based on the embedding of the encoded time series data from the time series encoder.

Abstract: A system and method for mitigating data drift in an industrial plant includes monitoring, by a processor, one or more process parameters associated with an industrial plant; detecting, by the processor, a drift in one or more process parameters based on a deviation from one or more predefined process parameters; determining, by the processor, one or more drift context and process context based on drift and one or more process parameters; determining, by the processor, sampling strategy from plurality of sampling strategies based on one or more drift and process context for sampling one or more process parameters using first Artificial Intelligence (AI) model; and training, by the processor, a second AI model based on sampling strategy for mitigating data drift.