Design News is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Google Launches Data-Rich Manufacturing Tools

Article-Google Launches Data-Rich Manufacturing Tools

Image courtesy of Alamy Feature.jpg
Google Cloud has developed cloud and edge systems specifically designed to help manufacturers improve operation efficiency.

The emergence of Google manufacturing tools was just a matter of time. Manufacturing has entered a world of massive data. Manufacturing equipment is generating tons of data that can be used to optimize performance. The data is processed at the edge and analyzed in the cloud. This data swirl is a big shift for manufacturers, but it’s normal life at Google.

The tools Google brings to manufacturers have been designed for the specific needs of automated production. “Google Cloud has launched new solutions that address many of the challenges manufacturers face as they strive to break down data silos, optimize machine throughput and availability, and ensure the highest product quality,” Charlie Sheridan, technical director of Industry Solutions–Manufacturing at Google Cloud, told Design News.

Google spent a few years learning manufacturing data while working with enterprise customers. Now the company is rolling out Manufacturing Data Engine and Manufacturing Connect for a wider market of manufacturers. “We’ve been active in manufacturing at the enterprise level and now we’re bringing it to the factory floor,” said Sheridan. “We’ve been active in this since we started Google Cloud.”

The tools are designed to help manufacturers connect siloed assets, process and standardize data, and improve visibility from the factory floor to the cloud. Once data is harmonized, Google’s tools help with three critical AI- and analytics-based use cases: manufacturing analytics and insights, predictive maintenance, and machine-level anomaly detection.

Manufacturing Data Engine and Manufacturing Connect are designed to unify data and empower the engineering workforce with easy-to-use analytics and AI solutions based on cloud infrastructure. “We’ve been active in manufacturing, now we're building package solutions so you can quickly deploy the technology,” said Sheridan. “The goal is to help improve manufacturing with analytics.”

Details on the Tools

Manufacturing Data Engine is an end-to-end solution that processes, contextualizes, and stores factory data on Google Cloud’s market-leading data platform. It provides a configurable and customizable blueprint for the ingestion, transformation, storage, and access to factory data. It integrates key Google Cloud products, including Cloud Dataflow, PubSub, BigQuery, Cloud Storage, Looker, Vertex AI, Apigee, and more, into a manufacturing-specific solution. 

Manufacturing Connect is a factory edge platform co-developed with Litmus Automation that quickly connects to, and streams data from, nearly any manufacturing asset and industrial system to Google Cloud, based on an extensive library of more than 250 machine protocols. Deep integration with the Manufacturing Data Engine unlocks rapid data intake into Google Cloud for processing machine and sensor data. The ability to deploy containerized applications and ML models to the edge enables new dimensions of use cases.

Image courtesy of Google CloudMDE 2 (002).png

This diagram shows the path for cloud analytics and edge processing.

The Goals of the Tools

Once data is centralized and harmonized by the Manufacturing Data Engine and Manufacturing Connect, it can then be used to address a growing set of industry-specific use cases, including:

Manufacturing analytics & insights were designed to help manufacturers quickly create custom dashboards to visualize key data—from factory KPIs such as Overall Equipment Effectiveness (OEE) to individual machine sensor data. Integrated with the Manufacturing Data Engine, engineers and plant managers can automatically set up new machines and factories, enabling standardized dashboards, KPIs, and on-demand drill-downs into the data to uncover new insights and opportunities throughout the factory. These can then be shared easily across the enterprise and with partners.

Machine-level anomaly detection was designed to help manufacturers identify anomalies as they occur and provides alerts—leveraging Google Cloud’s Time Series Insights API—on a real-time machine and sensor data such as noise, vibration, or temperature.

Predictive maintenance was designed to enable manufacturers to anticipate an asset’s need for service, helping reduce downtime and maintenance costs. Manufacturers can leverage ML models and high-accuracy AI optimizations that are deployable in weeks.

To determine the health of the equipment, Google uses data based on the equipment’s optimum performance.“We build models such as anomaly detection and deploy it to the edge We do the processing and create a virtuous system at the edge,” said Sheridan. “We do the processing at the edge and do the high-level analytics in the cloud. Get the data into the query and have our partners. We give free use cases and analytics, and optimization. We make it quick to deploy and easy for partners to get access to the data.”

Working with System Integrators

Google Cloud intends to work with manufacturers through their system integrators. We partner with system integrators and we will be adding more as we scale out. System integrators across the globe. We connected 100 machines to factories, managing 25 million records per week. Using system integrators is the best way to scale.

Google Cloud will not charge for the tools themselves. Instead, the company will charge for the use of the cloud and analytics processing. “There is no cost for the implementation code. It’s the cloud consumption and whatever analytics,” said Sheridan.

As for the return on investment for the manufacturer, that will come through factory efficiencies. “The cost savings come from better uptime and the longevity of the parts,” said Sheridan. “Once you nail predictive maintenance down, you can get high accuracy with the AI platform.”

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.