Sap Se

Lukas Carullo

Anubhav Bhatia

Patrick Brose

Leonard Brzezinski

Simon Lee

Lauren McMullen

Martin Weiss

Kun Bao

Karthik Mohan Mokashi

Rashmi Shetty B

Harpreet Singh

Markus Krabel

Srikanth Grandhe

Yogesh Beria

Emese Borbala Baliko

Daniel Huber

Sri Vidah A N

Provided is a system and method for training and validating models in a machine learning pipeline for failure mode analytics. The machine learning pipeline may include an unsupervised training phase, a validation phase and a supervised training and scoring phase. In one example, the method may include receiving an identification of a machine learning model, executing a machine learning pipeline comprising a plurality of services which train the machine learning model via at least one of an unsupervised learning process and a supervised learning process, the machine learning pipeline being controlled by an orchestration module that triggers ordered execution of the services, and storing the trained machine learning model output from the machine learning pipeline in a database associated with the machine learning pipeline.

Orchestrator for machine learning pipeline

Provided is a system and method for training and validating models in a machine learning pipeline for failure mode analytics. The machine learning pipeline may include an unsupervised training phase, a validation phase and a supervised training and scoring phase. In one example, the method may include receiving a request to create a machine learning model for failure mode detection associated with an asset, retrieving historical notification data of the asset, generating an unsupervised machine learning model via unsupervised learning on the historical notification data, wherein the unsupervised learning comprises identifying failure topics from text included in the historical notification data and mapping the identified failure topics to a plurality of predefined failure modes for the asset, and storing the generated unsupervised machine learning model via a storage device.

Failure mode analytics

Techniques and solutions are provided for processing data in conjunction with one or more hyperscale computing systems. An interface is provided for translating calls from an application into a format used by a hyperscale computing system. The calls can be to read data from, or write data to, a hyperscale computing system. In particular examples, data to be read or written is data from a plurality of IOT devices, where each IOT device has one or more hardware sensors. An interface can also be used to configure how the hyperscale computing system processes data, such as determining how IOT data is stored or how aggregates are generated from IOT data.

Interface for processing sensor data with hyperscale services

Techniques and solutions are provided for integrating master data from multiple applications. Master data from multiple applications can be integrated for use in processing data associated with internet of things (IOT) devices, such as by joining master data with timeseries data (including aggregated values). Integrating master data from multiple applications can include converting master data from a schema used by an application into an analytics schema. New or updated master data can be indicated in a message sent by an application. In processing the message, additional master data, or data used to determine how master data should be processed, can be retrieved.

Ingestion of master data from multiple applications

A data container for a content package comprising one or more semantics for populating the content package with selected types of information associated with a product or service is received by a computing device of a digital services framework. An organizational structure between and within networked tenants of the digital services framework is analyzed to identify one or more recipients for the content package. A data topology associated with the product or service is analyzed to generate announcements indicative of individualized content packages for the identified recipients for the content package. The announcements are sent to the identified recipients. Requests are received for subscriptions to the content package. Based on the analysis of the organizational structure and data topology and user-defined rules and semantics, instances of the container are selectively populating for tenants who have subscribed to the content package. The populated instances of the content package are sent to the subscribed users based on distribution data flows that are identified based at least in part on the analysis of the topological relationships and hierarchical structures.

Extraction and distribution of content packages in a digital services framework

A selection indicative of one or more types of information associated with a product or service is received by a computing device of a digital services framework. A data container is generated for a content package comprising one or more semantics for populating the content package with the selected types of information. Based on the product or service, one or more recipients for the content package are identified. An announcement indicative of the content package is sent to the identified recipients. Requests for subscriptions to the content package identified recipients are received. Instances of the container are populated for identified recipients who have subscribed to the content package. The populating comprises populating the content package with the selected types of information based on the one or more semantics. The populated instances of the content package are sent to the subscribed recipients.

Digital services framework

Provided is a system and method for predicting leading indicators for predicting occurrence of an event at a target asset. Rather than rely on traditional manufacturer-defined leading indicators for an asset, the examples herein predict leading indicators for a target asset based on actual operating conditions at the target asset. Accordingly, unanticipated operating conditions can be considered. In one example, the method may include receiving operating data of a target resource, the operating data being associated with previous occurrences of an event at the target resource, predicting one or more leading indicators of the event at the target resource based on the received operating data, each leading indicator comprising a variable and a threshold value for the variable, and outputting information about the one or more predicted leading indicators of the target resource for display via a user interface.

Growing community of inventors

Menlo Park, CA, United States of America

Lukas Carullo