Loading

FactoryTalk Analytics DataFlowML

Empower Your Data Scientists to Deploy Advanced Analytics Applications at Scale
Data scientist working at computer in control room

FactoryTalk Analytics DataFlowML

Empower Your Data Scientists to Deploy Advanced Analytics Applications at Scale
InnovationSuite
  • FactoryTalk Analytics
  • ThingWorx IIoT Platform
    • MES for Life Sciences
    • MES for Life Sciences
    • MES for Automotive
    • MES for Automotive
    • MES for Consumer Packaged Goods
    • MES for Consumer Packaged Goods
    • MES for Metals
    • MES for Metals
    • Alles bekijken
    • FactoryTalk Analytics DataFlowML
    • FactoryTalk Analytics DataFlowML
    • FactoryTalk Analytics Edge ML
    • FactoryTalk Analytics Edge ML
    • FactoryTalk Production
    • FactoryTalk Production
    • FactoryTalk Quality
    • FactoryTalk Quality

FactoryTalk® Analytics™ DataFlowML is a big data analytics and ML platform that empowers data scientists to visually build, train, deploy, score, and monitor ML models. With the explosion of large-scale industrial batch and streaming data in manufacturing, data scientists are constantly expected to build intelligent applications and uncover enterprise-level insights across site and plants. To ensure a faster route to innovation at enterprise scale, they need to be enabled to build and deploy analytics solutions, including machine learning models on big data—and reuse at multiple locations.

For this, data scientists need an open, standardized, secure, and enterprise-grade analytics platform on which they can collaboratively manage the entire model lifecycle in a centralized IT environment. While doing this, they should be able to blend and mashup data from multiple sources, leverage out-of-the-box open-source ML libraries for faster model building, and take advantage of pre-built ingress and egress data connectors to various IT or OT data sources.


Purpose Built for Data Scientists

This platform allows you to bring-your-own model, expedite building data pipelines using pre-configured ingress/egress options, and visually inspect data in motion. In addition, IT administrators can also leverage the platform to manage the lifecycle of data pipelines, thereby reducing total cost of solution. FactoryTalk Analytics DataFlowML is part of the FactoryTalk Analytics suite and integrates seamlessly with PTC® ThingWorx®, while offering connectivity to external applications. With FactoryTalk Analytics DataFlowML, data scientists can uncover the subtle, yet powerful operational insights, to drive business outcomes such as product quality enhancement, process optimization, and asset throughput improvement.

Loading

Pipeline graphic showing machine learning from FactoryTalk DataFlowML software

Visual Workflow-based Data Pipelines

Armed with visual data pipeline creation capabilities, you can create complex pipelines intuitively and apply machine learning algorithms—from Python, Spark, H2O.ai or PMML—on real-time streaming or batch data. Automatic schema detection and data inspection reduces development time, resulting in rapid prototyping. In addition, visualizing the transformed data at every step of the pipeline creation minimizes unexpected errors during execution.


Comprehensive Model Lifecycle Management

You can create build and train various ML models, incorporate them in data pipelines, and score them by deploying pipelines on a highly scalable Spark runtime execution engine. Model performance statistics and configurations are also available for visualization on in-built dashboards. The trained models can then be deployed via data pipelines to score against batch of streaming data at big data scale.

Screen captures from FactoryTalk DataFlowML software that show model configuration and metrics

Screen capture from FactoryTalk DataFlowML software that show icons for available out-of-the-box

Out-of-the-box Ingress, Egress, Transformation & Analytical Processors

The platform has out-of-the-box connection adapters, data transformation processors, and emitters that can be visually configured. You can enable access to data from various sources such as message queues, transactional databases, log files, and controllers. The egress of a pipeline could be MySQL, queues, relational databases, third-party BI tool, cloud storage containers, or ThingWorx. This drastically reduces design time and enables faster iterations


Open Programmable Platform

Bring in your datasets, pre-built machine learning or analytical models, and import code built on a programming environment of your choice such as Python or Jupyter. This allows you to scale building analytics applications and increase the pace of innovation by maximize reusability of existing assets across the enterprise.

Screen captures from FactoryTalk DataFlowML software that shows programming code being used in Python and Jupyter platforms

Ready to Order Your Software?

Ordering Options
Software SubscriptionsSoftware Subscriptions
Ordering Options

One stop shopping for software and subscription services.

Explore Our Ordering Options
Talk to a Consultant
DistributorDistributor
Talk to a Consultant

Work with your local Rockwell Automation sales consultant to discuss your software requirements.

Contact Us

Recommended For You

Loading
Loading
Loading
  • Technische vraag
  • Technische ondersteuning via chat
  • Technische ondersteuning via telefoon
  • Neem contact op met Verkoop
  1. Chevron LeftChevron Left Nederland Chevron RightChevron Right
  2. Chevron LeftChevron Left Pro... Chevron RightChevron Right
  3. Chevron LeftChevron Left Software Chevron RightChevron Right
  4. Chevron LeftChevron Left FactoryTalk InnovationSuite Chevron RightChevron Right
  5. Chevron LeftChevron Left FactoryTalk Analytics DataFlowML Chevron RightChevron Right