Put Big Data into context in real time

Sheila Kennedy says make more-strategic decisions thanks to the power of AI and machine learning.

By Sheila Kennedy, contributing editor

Smart cities, plants, and machines rely on smart analytics to put big data into context in real time and optimize decision-making. Powering today’s analytics tools are advanced algorithms, artificial intelligence (AI), machine learning, and digital twins that pull disorganized, disconnected, and “dark” (captured but unused) data into the mix of available information and make it actionable.

Elevating industrial intelligence


Strategic algorithms and cognitive computing improve operational acuity. Seeq software uses end-user-focused advanced computing algorithms to collect and analyze time-series data from manufacturing systems and process historians in real time. Trending, pattern and limit searches, outlier detection, modeling, and illustrative visualizations are among the capabilities that can be applied directly to the data sources without assistance from data scientists or IT experts.

Seeq customer engineers and experts possess “a gold mine of knowledge and know-how,” says Brian Parsonnet, chief technology officer and founder at Seeq. “The software automates low-level procedural software tasks, empowering users with expertise in plants and processes to leverage their experience to accelerate insights.”

AI-powered predictive analytics software from Canvass Analytics automates the entire data analysis process and creates adaptive, predictive data models. It allows operations teams to “achieve real improvements in asset uptime through predictive maintenance programs, increase energy efficiency, improve quality, and optimize production processes, knowing they have the latest insight from their connected factory floor,” says Humera Malik, CEO of Canvass Analytics.

She explains: “AI-powered predictive analytics can help plant operations teams reveal hidden insights into the millions of data points that their connected machines are generating and then go a step further by recognizing when changes occur to the operating environment and updating these insights in real time.”

Integrated MCare from L&T Technology Services is an end-to-end condition-based maintenance solution that captures data through industrial edge hardware and sensors and analyzes it with the help of incremental AI algorithms, both at the edge and on-server. “With the increasing complexity of modern factory equipment, reactive maintenance practices can bring down a plant’s overall productivity,” notes Ashok Kumar, head of digital and industrial solutions at L&T Technology Services.

Integrated MCare “extracts signatures from a given set of data, which helps in detecting equipment failures at an early stage through continuous monitoring,” Kumar says. “This reduces the cost of unnecessary maintenance and expenses of unplanned downtime.”

With tools such as ABB Ability Enterprise Analytics, producer-supplier collaboration is improved because both parties see the same thing at the same time and can resolve issues more quickly, or predict and avoid them. With ABB Ability’s cloud-based data architecture, this is done on a massive scale, says Kevin Starr, global advanced digital services program manager for ABB’s oil, gas and chemicals business.

“We apply repeatable, automatic data-gathering and analysis methodologies and easy-to-understand data analytics for multiple assets, processes, and risks,” says Starr. “These include control systems, control loops, cybersecurity, drives, quality control, rotating machines, and more.” This family of “analytics apps” has machine-learning features so that thresholds for predictive notifications are set with pinpoint accuracy for real-time mitigation, he adds.

Knowledge through virtualization


Digital twin capabilities allow analytics to be formed from a virtual mirror of assets or processes. The Element Platform from Element Analytics is a digital-twin management solution that “unlocks ad hoc analysis on any and all operational data,” says Sameer Kalwani, founder and vice president of product for Element Analytics.

Digital twins allow companies to compare assets, predict equipment failures, and optimize process lines using simple business intelligence tools or advanced machine learning, explains Kalwani. “Companies are realizing that building, let alone maintaining, digital twins requires software,” he adds.

The Digital Twin Builder (Advanced) within Sight Machine’s Enterprise Manufacturing Analytics (EMA) solution enables configuration of a unified data model of the enterprise. Sight Machine’s digital manufacturing platform provides “a scalable pipeline to combine factory data into standard manufacturing data models that mirror machines/lines, parts/batches, plants, and supply chains,” says Ryan Smith, vice president of product and engineering at Sight Machine.

Another EMA feature is its Correlation Heatmap tool, which automatically analyzes tens of thousands of data points to identify which variables and machines are affecting a product’s quality. “Using advanced analytics and machine learning, Sight Machine’s platform helps companies increase productivity, improve quality, and provide remote visibility across the manufacturing enterprise,” says Smith.

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments