Current Issue


Technology ADAPT Reports

Building an Intelligent Manufacturing Platform

Order Reprints

Dec. 28, 2020—Earlier this month, industry professionals gathered to discuss what steps it would take to build an intelligent manufacturing platform. Automation is on the rise, and with it, comes artificial intelligence, but there are many building blocks that must be scaled in order to have a fully-functional, intelligent manufacturing plant. 

Michael Ger, the managing director for manufacturing and automotive at Cloudera, and Brian Irwin, automotive and industrial leader for Accenture, detailed the processes and life cycles behind intelligent manufacturing during the Center for Automotive Research’s most recent webinar. 

ADAPT tuned in and took notes to bring you this guide so your facility can stay at the cutting edge, rather than playing catch up down the line. 

Machine-Learning Life Cycle

To kick off the webinar, Ger began by describing the life cycle of machine-learning and analytics. He said everything begins at the manufacturing plant where the data is collected. The data from the plant then gets processed and transmitted for review. Once the data is ingested and analyzed, it moves into the cloud for storage. 

From there, Ger said the data enters the “enterprise” phase, which involves analyzing the data for self-service business intelligence opportunities and other enterprise analytics. 

After the data has completed each of the analyses, begins the learning process for the machines. The machine’s use of the data is then monitored for potential hiccups before it is officially deployed to be used in real-life processes.

“Data is used to provide greater levels of automation and enable processes like predictive maintenance,” said Ger. The goal when using machine-learning is to cut down on cycle times, increase efficiency, and eventually reduce repair severity. 

Building Blocks 

Irwin said, “Organizations today are rich in data, but poor in insight,” so he crafted a set of building blocks that he said must be in place in order to have a functioning, intelligent platform for advanced manufacturing.




Small steps to work toward:

Connected worker: digital technologies such as wearable devices should be used to assist shop floor workers in executing operational activities 


Control tower: the plant should have a digitally-enabled control tower to assist in overseeing manufacturing processes and network operations management 


Cyber security: threat prevention techniques should be put in place to predict, protect, and understand operational risks across IT platforms 


Predictive maintenance: predictive analytics for maintenance should be put in place to reduce asset downtime and increase revenue


Digital safety: safety risks should be reduced by monitoring an individual’s location and exposure to hazardous materials


Automated decision-making: artificial intelligence and machine-learning should propose actions by monitoring production volume, safety, and quality 


Connected lines: data should be integrated from multiple sources and should be channeled to the relevant user interfaces


Digital quality: the facility should utilize connected systems and artificial intelligence to identify slight changes in quality, minimizing rework time


Irwin said Accenture’s approach to intelligent manufacturing is to think big, but start small. If your facility wants to be at the cutting edge of machine-learning, he recommends you do the same. 

“You won’t be able to tackle it all at once, so start with a piece of it,” he said.  


Related Articles

ZETA’s Building Blocks for an All-Electric 2030

BlackBerry Partners With Amazon for Vehicle Platform

GM Building EV Factory in Quebec

You must login or register in order to post a comment.