Abstract: An anomaly diagnosis system obtains a plurality of anomaly signals corresponding to a plurality of sensor signals of a physical system and segments one or more anomaly signals into a plurality of time segments. The system determines an anomaly score for each time segment based on anomaly values of the one or more anomaly signals during the time segment and identifies an anomaly time interval corresponding to at least one consecutive time segment within the plurality of time segments. The system clusters the plurality of anomaly signals within the anomaly time interval to identify an anomaly group of sensor signals associated with the anomaly time interval and determines an aggregate anomaly score for the anomaly group. The system generates a graphical user interface presenting a representation of the anomaly group of sensor signals and the aggregate anomaly scores and causes the graphical user interface to be displayed on a user device.

Abstract: A computer system for managing a machine learning model that detects potential anomalies in the operation of a complex system is disclosed. In some embodiments, the computer system is programmed to receive sensor signal data originally produced by sensors of the complex system. The sensor signal data can include values for multiple sensor signals at multiple resolutions. The computer system is programmed to train, from given sensor signal data, the machine learning model that comprises one or more transformers, each transformer capturing a set of relationships between signals in a predetermined group of signals. During training, the computer system is programmed to also establish an expected range for an indicator of the relationship. The computer system is programmed to then execute the machine learning model on new sensor signal data and take remedial steps when any computed indicator falls outside the expected range, indicating a potential anomaly in the operation of the complex system.

Abstract: A method comprises generating, from time series data, a plurality of tiles for each resolution of a plurality of resolutions, a first plurality of tiles associated with one resolution covering the same time period as a second plurality of tiles associated with another resolution, each tile having a common number of N values representing all measurements associated with a duration of time covered by the tile in the time series data; receiving a first user request specifying a first timestamp and a first resolution; determining that no tile is available based on the first timestamp and the first resolution; generating a first tile covering a first duration of time based on the first timestamp and the first resolution, a first number of measurements associated with the first duration of time being less than a second number of measurements associated with a second duration of time based on the first resolution; transmitting the first tile.

Abstract: A tile contains aggregated data at a certain resolution for the actual data present in the duration of time covered in that tile. Tiles are generated at every possible resolution suitable for a computer display and provide aggregate measures such as averages and variances. Tiles provide a summary of data such that from the highest level view down to the specific data time points collected, a user's attention may be drawn to the times when there is the most interesting pattern behavior to review and analyze.

Abstract: Techniques for managing machine operations using encoded multi-scale time series data are provided. In one technique, operational data is received from a sensor coupled to an industrial device. For each portion in a first set of portions of the operational data (where each portion corresponds to a first time scale), first aggregated data is generated based on time series data from that portion and a first encoding is generated based on the first aggregated data. For each portion in a second set of portions of the operational data (where each portion of the second set corresponds to a second time scale that is different than the first time scale), second aggregated data is generated based on time series data from that portion and a second encoding is generated based on the second aggregated data. The operational data is classified to determine a condition of the industrial device during the time interval based on the first and second encodings.

Abstract: Model chaining provides users with enormous flexibility to define their systems in a way that best suits their needs to get the most benefit from artificial intelligence models. In model chaining, a model chain may be generated. Output of a model is used as the signal input to another model. In this way, lower-level models can be more sensitive as they find patterns using just a few signals, and higher-level model then looks for patterns in the patterns of the lower-level models. All of the signals are used while users are not being blinded by more subtle behaviors.

Abstract: A computer-implemented method of predicting an event horizon is disclosed. The method comprises maintaining condition data indicating a plurality of conditions occurring on one or more physical systems at a plurality of points in time. The method further comprises receiving an input feature vector representing a given condition occurring at a specific time during a specific period of time. The method also comprises generating, using a particular trained machine learning model of a plurality of trained machine learning models, a forecast value that indicates an amount of time from the specific time to an occurrence of a particular target condition on a particular physical system, the particular target condition being different from the given condition, each trained machine learning model corresponding to a distinct target condition. In addition, the method comprises causing, based on the forecast value, an action to be executed on the particular physical system.

Abstract: A method for generating forecast predictions that indicate an event horizon of an entity or remaining useful life of a consumable using machine learning techniques is provided. Using a server computer system, feature data comprising features vectors that represent a set of signal data over a range of time is stored. Condition data comprising conditions occurring on the entity at particular moments in time is stored. Label data that comprises a plurality of time values that each indicate a difference in time between one condition and another condition is stored. A training dataset is created by combining the feature data, the condition data, and the label data into a single dataset. The training dataset is partitioned by condition. A machine learning model is trained on each target condition training dataset. The trained machine learning models are used to generate forecast values that each indicate an amount of time to an occurrence of a target condition associated with an entity.

Abstract: A tile contains aggregated data at a certain resolution for the actual data present in the duration of time covered in that tile. Tiles are generated at every possible resolution suitable for a computer display and provide aggregate measures such as averages and variances. Tiles provide a summary of data such that from the highest level view down to the specific data time points collected, a user's attention may be drawn to the times when there is the most interesting pattern behavior to review and analyze.

Abstract: A method for generating forecast predictions that indicate an event horizon of an entity or remaining useful life of a consumable using machine learning techniques is provided. Using a server computer system, feature data comprising features vectors that represent a set of signal data over a range of time is stored. Condition data comprising conditions occurring on the entity at particular moments in time is stored. Label data that comprises a plurality of time values that each indicate a difference in time between one condition and another condition is stored. A training dataset is created by combining the feature data, the condition data, and the label data into a single dataset. The training dataset is partitioned by condition. A machine learning model is trained on each target condition training dataset. The trained machine learning models are used to generate forecast values that each indicate an amount of time to an occurrence of a target condition associated with an entity.

Abstract: A tile contains aggregated data at a certain resolution for the actual data present in the duration of time covered in that tile. Tiles are generated at every possible resolution suitable for a computer display and provide aggregate measures such as averages and variances. Tiles provide a summary of data such that from the highest level view down to the specific data time points collected, a user's attention may be drawn to the times when there is the most interesting pattern behavior to review and analyze.

Abstract: A tile contains aggregated data at a certain resolution for the actual data present in the duration of time covered in that tile. Tiles are generated at every possible resolution suitable for a computer display and provide aggregate measures such as averages and variances. Tiles provide a summary of data such that from the highest level view down to the specific data time points collected, a user's attention may be drawn to the times when there is the most interesting pattern behavior to review and analyze.

Abstract: A method is provided. The method stores a first version of an artifact in a first environment. Further, the method stores a second version of the artifact in the first environment. In addition, the second version of the artifact is distinct from the first version of the artifact. The method also deploys the first version of the artifact and the second version of the artifact to a second environment so that the first artifact and the second artifact can be run simultaneously in the second environment. The second environment is distinct from the first environment.

Abstract: A method is provided. The method develops code for an enterprise module in an enterprise module development environment. Further, the method modifies the code to output customized code that runs in an enterprise module production environment. In addition, the method deploys the customized code to the enterprise module production environment.

Abstract: A method is provided. The method composes management code in a platform independent managed object format. Further, the management code is utilized to manage at least one object. In addition, the management code transforms the management code into a plurality of portable management objects and a plurality of native management objects.

Abstract: A method is provided. The method determines a set of components to include in an enterprise module. Further, the set of components is divided into a set of platform dependent components and a set of platform independent components. In addition, abstract computer code is prepared for each of the components in the set of platform dependent components according to at least one of a plurality of high-level abstract computer languages. Further, the abstract computer code is provided to a transmogrifier to automatically generate platform dependent source code. In addition, the platform independent source code is prepared for the set of platform independent components. The enterprise object code is generated by compiling and linking the platform dependent source code and the platform independent source code.

Abstract: A method and code generator system for integrating different enterprise business applications is disclosed. In one embodiment, a method for integrating a local business system with an external business system, comprises using a code generator to generate integration source code, wherein using a code generator comprises; interrogating a repository containing integration data by an introspector; and using the integration data with a code filter, wherein the filter generates the integration source code.

Abstract: A method is provided. The method composes management code in a platform independent managed object format. Further, the management code is utilized to manage at least one object. In addition, the management code transforms the management code into a plurality of portable management objects and a plurality of native management objects.

Abstract: A method is provided. The method designates an intersection library that includes an intersection of a first set of pre-constructed code that utilizes a first set of syntax language from a first platform and a second set of pre-constructed code that utilizes a second set of syntax language from a second platform. Further, the method composes platform independent code for an application service for an enterprise module. The platform independent code is compiled and linked with the intersection library. In addition, the method designates a portable library that includes a third set of pre-constructed code that utilizes a third set of syntax language from a third platform and a fourth set of pre-constructed code that utilizes a fourth set of syntax language from a fourth platform. Further, the third set of pre-constructed code and the fourth set of pre-constructed code are composed in order to accomplish the same task.

Abstract: A method is provided. The method determines a set of components to include in an enterprise module. Further, the set of components is divided into a set of platform dependent components and a set of platform independent components. In addition, abstract computer code is prepared for each of the components in the set of platform dependent components according to at least one of a plurality of high-level abstract computer languages. Further, the abstract computer code is provided to a transmogrifier to automatically generate platform dependent source code. In addition, the platform independent source code is prepared for the set of platform independent components. The enterprise object code is generated by compiling and linking the platform dependent source code and the platform independent source code.