New Platforms Aim to Scale IIoT Analytics

July 31, 2018
As manufacturers struggle with data acquisition across multiple plants, Sight Machine responds with tools that collect, transform and transmit data from disparate devices and data lakes.

In the age of the Industrial Internet of Things (IIoT), many manufacturers are sold on the idea of combining edge and cloud analytics to make sense of plant floor data in order to make better production decisions. And, there have been many pilot projects to date that test the analytics abilities on a machine in a facility. It’s worked. And the plant managers and senior executives are thrilled.

Now, these companies have a new challenge: How to scale the operation to an entire production process as well as to multiple facilities around the globe.The problem they face is that they have to tap into a variety of dissimilar assets, data types, vendors and systems—from programmable logic controllers (PLCs) to enterprise resource planning (ERP). They also must work in geographies that don’t have reliable Internet access.

There are a number of new analytics platforms that can help manufacturers sift through the mountains of data. One such company, Sight Machine, built an open source model that uses artificial intelligence (AI) and machine learning algorithms to work with structured and unstructured data. The company’s tools also define a plant floor digital twin, configured to mirror manufacturing production, enabling the AI to convert and contextualize data and machine learning to identify anomalies.

Sight Machine's Enterprise Manufacturing Analytics (EMA) platform has worked well for its customers, company officials report. Recently, however, the technology supplier recognized its customers' need for speed, simplicity and scalability and, as a result, rolled out two new products on the platform to address the issues.

In June, Sight Machine introduced FactoryTX Edge and FactoryTX Cloud, both of which enable rapid deployment and centralized management of multi-factory IIoT data ingestion. According to Sight Machine, FactoryTX Edge and Cloud give manufacturers the flexibility to acquire machine data wherever it exists—directly from machines, from disparate devices, local data storage or from cloud-based data lakes—pre-process it into a standardized format and securely stream it to Sight Machine’s cloud-based analytics software.

FactoryTX Edge acquires raw data from production environments while addressing network reliability challenges through micro batching and software update delta compression, using store-and-forward capabilities to mitigate bandwidth limitations in a plant. It includes self-service installation technology that allows end users to remotely deploy digital manufacturing applications, and enables remote provisioning, monitoring and maintenance to accelerate the ability to scale and improve digital manufacturing capabilities via browser-based configuration tools. Factory TX Cloud offers the same functionality as Edge. But it is optimized to run in the cloud, enabling manufacturers to quickly turn the aggregated data into informational insights.

In addition, the new tools are designed to mesh with enterprise-wide information technology (IT) imperatives to avoid creating new data silos and isolated local applications.

“We help the manufacturer acquire the data off the factory floor, create a digital twin and then use models to help analyze the business,” said Ed Jimenez, vice president of marketing at Sight Machine. “Factory Edge and Cloud are in that first bucket. It’s an optimized way to acquire data off the factory floor and to do it at scale." It doesn’t replace the things manufacturers are already doing to capture data. It just enables the collection of information from a wider variety of systems, he said.

For example, Sight Machine worked with a large paper packaging manufacturer challenged by high scrap rates in its production mills. As a global manufacturer, the company was focused on building a scalable platform to analyze a wide range of production data to address quality challenges.

The company had a historian at each facility capturing information from hundreds of PID controllers used throughout production. The data was forwarded to a cloud-based data lake for centralized storage of data from multiple facilities. But there was no easy way to see relationships between that data. The manufacturer wanted an IoT analytics platform that continuously analyzed all data in real time rather than manually-selected historical samples, which would allow them to investigate and react faster.

So, it implemented Sight Machine’s Enterprise Manufacturing Analytics application suite to analyze thousands of sensor readings for every machine that touched individual paper rolls throughout the production process. The manufacturer used this analysis to rapidly identify hidden relationships between production parameters and quality issues to reduce scrap rates and improve quality and yield.

“We are excited to see an increase of customers driving toward operational scale,” said Ryan Smith, Sight Machine’s vice president of engineering. “We had wins at the plant level, but companies are now thinking of multiple lines in a plant or multiple plants in manufacturing or the supply chain network. The consideration now is how we enable customers to do this faster and easier.” That’s where FactoryTX comes in. “FactoryTX is scaling data to add business value.”

Companies in this Article