Applying advanced analytics and machine learning to reduce operational costs and risks
Predix Platform provides a rich industrial-grade analytics library and framework to create machine learning analytics tailored to digital twin asset models
It's not enough for today's industrial organizations to just make data-driven decisions, they must also empower data to help automate business. With a lot of industry buzz around applied analytics and machine learning it can be difficult to strategize a roadmap of powerful analytic applications to drive specific business outcomes-Predix Platform can help. Predix Platform not only provides a rich set of industrial-grade analytic libraries but it also provides a unique framework to create machine learning analytics. These analytics assist organizations in predicting critical failures, automating process, and driving efficiencies. When applied to real-time Industrial Internet data streams, edge- or cloud-based analytics can detect anomalies, triage critical events, direct prescriptive controls, signal predictive maintenance alerts, and more.
Choose from pre-built industrial analytics tested by GE domain experts across our industrial businesses.
Easily build, test, and deploy machine learning models.
Leverage in-house expertise as well as open source algorithms and tools.
Deploy self-learning analytics to automate processes at the edge or in the cloud.
GE also offers data science services to complement its analytics technology. This team offers flexible, rapid engagement packages to help customers analyze data, set analytic objectives, and develop customer analytics that can be deployed on Predix Platform.
A set of over 100 algorithms and models is available in Predix Platform Analytics Marketplace. A finished analytic written in C++, Java, or Python can be orchestrated to run based on a time schedule or event occurrence. Alternatively, complex analytics developed outside Predix Platform using the preferred tools of a data scientist, can just as easily be incorporated into your application. Base algorithms can be sourced from open source libraries and configured in your Predix Platform powered applications. Once incorporated your Predix Platform powered application can use the results of the analytic for a number of purposes from creating alerts on outages or signaling unplanned downtime.
Not all data is created equal, some data is critical to analyze as close as possible to the asset and others can be processed at a gateway or in the data center. Use Predix Machine to execute analytics as part of local data processing to filter your critical data points. An analytic can be created using the preferred tools of a data scientist and then packaged as a container whose execution can be orchestrated by the edge manager in the same way as any other edge application.
GE has a long history of analytics development centered around GE-manufactured assets and GE-delivered services. Predix Asset Performance Management (Predix APM) offers a rich analytics catalog purpose-built for specific asset reliability monitoring. These models can be trained and deployed from within Predix Asset Performance Management. Alternatively, these analytics may be included in products or services offered by specific GE businesses.