Challenges with the Traditional Approach to Optimization
Organizations face difficult challenges in their analytics adoption and maturity progression. The first is around data access with the surrounding challenges of getting the right context, understanding the data relationships and processing large volumes of data for more advanced analytics.
Second, is trust in the insights due to poor data quality, not enough training of data and lack of domain knowledge to achieve accurate results. There’s also the challenge of an inability to scale the apps due to the lack of repeatability, reusability and complications in orchestrating and maintaining large scale deployments.
And finally, the challenge of siloed visibility and control due to disconnected apps that focus only on specific areas, and the apps are hard to combine.
Rockwell Automation addresses these key challenges by providing an industrial data, AI and edge platform and delivering key analytics applications that address the top industrial use cases.
The Impact of Data and Analytics on Decision-Making
Manufacturers believe that the challenges are becoming more complex. According to Top Trends in Data and Analytics, 2022 Gartner, 47% of organizations believe that the decisions they face will be more complex, increasing demand for connected, continuous and contextual data and analytics and explainable decision processes.
There are many analytic technologies in the market but most require deep expertise in both data science and industrial processes which often prevents operational technology (OT) professionals from taking advantage of them. FactoryTalk Analytics LogixAI software continuously monitors the production operation and makes predictions about your process, enabling operators and technicians to easily apply machine learning concepts to get ahead of product quality issues and help protect process integrity—all without having to develop, validate or sustain home-grown machine learning algorithms.
“FactoryTalk Analytics LogixAI dials in on prescriptive analytics to determine projections about how your plant will perform specifically through Soft Sensors® or software that uses real-time production data and models to determine what is going on in your facility to make predictions about one or more KPIs that you’re interested in," Amanda said. “The software makes the data available sooner and faster and also available in real-time versus lab analysis to understand what is influencing the variable.”
FactoryTalk Analytics LogixAI software maximizes asset performance with continuous monitoring, real time data analysis and prediction of a single target outcome. It’s edge-based and empowers OT personnel with a physics-based model that makes a logical prediction of an operational parameter’s value.
The module runs in the chassis and communicates with ControlLogix® via the backplane or the containerized app, runs on an industrial edge PC and communicates via Ethernet.
Perfect Fill
When it comes to fluid products, the filling or dispensing step is one of the most critical and expensive stages in the process. But many manufacturers struggle to achieve accurate and consistent filling which results in over- or under-filling. Overfilling results in unnecessary product giveaway, higher costs and lost revenue, while underfilling can increase product and packaging scrap, re-work and regulatory and consumer risk. FactoryTalk Analytics LogixAI software analyzes the available production data and builds a model to predict the weight of product being filled in each package. By deploying the model at the edge, the software makes real time predictions based on current operating conditions. These predictions help the operations team control the filling process, giving them information they need to increase the fill setpoint when the weight is predicted to be low and decrease the fill setpoint when the weight is predicted to be high.
Perfect Fit
The term “splice” is used to describe the length of overlapping material where a material bonds. For industrial processes with rolled products that involve splicing, the splice length is a key process indicator. Short splices reduce product quality but long splices waste raw material. Often, operators need to make manual adjustments to the process to achieve consistent, in-tolerance splice lengths. With FactoryTalk Analytics LogixAI software, you can reduce manual operator interaction by using live or historical data to build and train a machine learning model that predicts whether splices will be in or out of tolerance. Predictions can be integrated with ControlLogix to consistently make process adjustments that proactively correct out of tolerance splices, increasing productivity and profitability.
Now Supporting CompactLogix and FactoryTalk Logix Echo
FactoryTalk Analytics LogixAI software version 3.1 now supports integration with CompactLogix 5380® controllers. With this release, automation engineers using CompactLogix 5380 to control their production can deploy no-code Soft Sensor models at the edge. The software uses the controller tags as the data source and transforms streaming controller data into calculated predictions at the high speeds required for process control. The predictions can be implemented by operations teams in place of a manual reading or in a place where it’s not possible to deploy a traditional sensor to improve quality, increase yield, reduce variability or increase asset utilization.
In addition, the new version 3.1 supports integration with FactoryTalk Logix Echo controller emulation software. Automation engineers can now quickly and easily build FactoryTalk Analytics LogixAI predictions into their emulation systems for testing and training in a virtual environment.
Reduce manufacturing scrap, minimize product giveaway and increases plant profitability with FactoryTalk Analytics LogixAI software. Learn more at FactoryTalk Analytics LogixAI and watch the on-demand webinar.
