Unexpected downtime happens during operations, and it affects overall customer experience. The solution empowers solution builders to create applications that help maintenance personnel to perform just-in-time maintenance programs that reduce unexpected downtime and improve customer experience. Use this packaged analytic to enable “what-if” scenarios based on subtle changes in operating conditions. The goal is to find an acceptable operating pattern that yields reduced wear and tear on the asset, thus extending its remaining useful life and reducing failure rates. Use it to lower repair costs and repair frequency and increase overall maintenance predictability.
The ML services model manager trains your specific ML model by accessing historic data and routing datasets through the ML service to influence responses. It can optionally call third-party simulation engines, sending payloads, and receiving back the result, synchronously. This creates the ability to augment existing sensor data with inferred values that can’t be easily measured in the physical world, such as angular acceleration, torque, and so forth. The embedded asset model also provides the user the ability to create, update, and delete a digital asset from the user interface.
In deployment, data ingestion provides the ability to read data from industrial sources and sensors. Data processing is applied to the ingested data by reading its payload and redirecting parts of it to select destinations. Internal data storage functions are included with the ability to also forward the collected data to a destination of choice. Data can be exchanged with digital twins with the ability to create, update, and delete digital assets as needed. Prebuilt ML models can be applied as analytics applications based on customer-specific operational challenges and requirements.
The same simulation capabilities can be extended to run-time, where the digital twin is constantly updated and compared with calculated results to correct for asset degradation: for example, motor-bearing wear or heat-exchanger fowling. That way the analytics can account for the state of your equipment at any time during its life cycle.
Analytics can be deployed to on-premises and cloud platforms.
As a step up from the Anomaly Detection and Prediction Analytic and the applications developed with it, APM and Lumada Manufacturing Insights industry solutions are designed to provide health and performance insights to prevent critical asset failures while optimizing asset life-cycle costs. These solutions enable companies to leverage their online and offline data to drive more intelligent, risk-based approaches to asset management in alignment with industry standards, such as ISO 55000 and PAS 55.
The capabilities that Hitachi provides through Lumada’s software for IIoT and Lumada DataOps extend your data management and analytics framework into industrial operations.
Hitachi Vantara solves digital challenges by guiding you from what’s now to what’s next. Our unmatched industrial and digital capabilities benefit both business and society