Streaming data – data that is generated continuously at the frequency of seconds or even faster; has become the corner stone of industrial IOT helping manufacturing companies in their journey to Industry 4.0. Real time alerts and notifications based on streaming data have been well established in both discreet and process manufacturing setups. Such alerts have become the cornerstone of predictive maintenance, predictive quality, safety management and environment management solutions.
In the next phase of evolution, enterprises are putting in place complex rules and machine learning models that are being applied on such streaming data. In addition, enterprises are also looking for ways of summarizing and historizing such data for future analysis and model building.
While real time data is being used for streaming dashboards, enterprises are also interested in combining that data with less dynamic process data from Manufacturing Execution Systems and ERP Systems to extract combined insights and apply machine learning models based on process data and streaming data.
Next generation IOT platforms are gearing up for these challenges to help enterprises take the next leap in predictive maintenance, predictive quality, safety management and environment management. And thereby help embrace the Industry 4.0 philosophy.
Common problems:
Difficulty in incorporating machine learning models on streaming data.
Inability to combine streaming data with process data from various other systems like MES, ERP etc.
Difficulty in obtaining analytical data for model building.
Limited flexibility in summarizing streaming data – summarizing window selection, methods of summarization (mean, max, min, stdev, no of times above or below threshold, kurtosis etc.)
Limited flexibility in incorporating control charts for creating alerts and notifications.
LeoStream, the end-to-end streaming data management solution
Lambda Architecture
Lambda architecture to address dual needs of streaming and summarized dashboards.
Drag-and-Drop
Drag-and-drop wizard provides complete flexibility to summarize streaming data in any way the user wants.
Analytical Data Store
Analytical data store that can be used by data scientists to get model ready data.
Easy Interface
Easy interfaced to implement machine learning models.
Cloud Ready
Cloud ready solution.