Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
Three Best Practices for Unilever’s Global Analytics Initiatives
    This article from Morgan Vawter, Global Vice...
Getting Machine Learning Projects from Idea to Execution
 Originally published in Harvard Business Review Machine learning might...
Eric Siegel on Bloomberg Businessweek
  Listen to Eric Siegel, former Columbia University Professor,...
Effective Machine Learning Needs Leadership — Not AI Hype
 Originally published in BigThink, Feb 12, 2024.  Excerpted from The...
SHARE THIS:

This excerpt is from diginomica. To view the whole article click here

9 years ago
Big Data is Helping Manufacturers Think Bigger

 

Everyone is running around talking about big data nowadays – and yes, our databases are immense. A gigabyte used to sound like a lot. Today, we talk about terabytes without blinking.

But big data didn’t grow up overnight, just as that high school graduate didn’t suddenly wake up as a teenager with an iPhone and driver’s license. Data volumes have grown steadily for many years. But today, the value we get from all those ones and zeroes is increasing exponentially.

Big Data Meetup

This is as true in manufacturing as it is in any other industry. Here are three ways big data is helping manufacturers think bigger than ever before.

1. Monitoring product quality proactively

Cloud computing. Cloud data storage. The Internet of Things. All these factors have collided to create a golden opportunity for manufacturers, who can now use massive data volumes in unexpected new ways.

In a recent Plex survey of manufacturers, 66 percent of respondents said cloud technologies have already improved insights into their business. Some 31 percent are either evaluating their big data needs and opportunities or plan to do so in the coming year – and 35 percent believe data analysis is the most important skill set for the next generation of employees.

All this data enables manufacturers to keep a much closer eye – or ear – on product quality. Just ask our customer Inteva Products, a major automotive component maker. Inteva doesn’t just inspect their sunroof motors visually – they actually listen to them run to make sure they’re not too loud. The company stores thousands of audio files of these motor tests each day. As storage and analysis technologies develop, Inteva will gain insights that enhance their future product development – insights their competitors lack.

Big data doesn’t just reassure manufacturers that they’re producing high-quality products – it also convinces their customers. Manufacturers now provide an incredible breadth and depth of data on their products’ construction and testing, establishing up front that they’ve delivered something of lasting value. Most buyers would agree that hard data beats marketing hype any day of the week.

Stop and think about where these developments are taking us. Manufacturers will soon be able to eliminate statistical process control from their quality control process. Instead, they’ll use today’s increasingly affordable sensors to gather real-time data on every item that comes off the assembly line. I’m talking about dozens of measurements on each product. The bottom line will be greater accuracy with less human involvement.

2. Seeing the future – and changing it

Plex-manufacturing-Plant-floor-operationsOperational analytics are great at telling us what just happened and why. Manufacturers have been doing that kind of analysis for years. But they’re now using the predictive aspects of big data to monitor their operations against their quality standards. That often means predicting when a machine or tool is about to break – before it starts churning out defective products.

Predictive analytics tell us what’s about to happen. Prescriptive analytics show us how to make machines do what we want. These disciplines are the crown jewels of business intelligence. Both require vast amounts of data – and the ability to analyze it effectively. That’s what big data delivers for today’s manufacturers.

It’s one thing to look at the history of maintenance issues or failures on a specific machine. But today, manufacturers factor in so many other variables. They’re looking at a press and factoring in not only all the metrics around it – such as temperature and tonnage – but also who’s working on the machine, how long of a shift they’ve worked, what tools are in the press, the day and time of year, and much more. They then factor all of this data into their predictions of when future failures will occur.

Predictive analytics isn’t a new discipline. But until recently, its high cost made it practical for only very expensive products or shop equipment. New tools are making predictive analytics a way of life for manufacturers of all sizes. And as the Internet of Things continues to mature, manufacturers are gathering more and more data automatically.

Image credit: Welding robots in a car factory © taaee – Fotolia.com; workshop scene by Plex.

By Jerry Foster
Originally published at http://diginomica.com

This excerpt is from diginomica. To view the whole article click here.

Leave a Reply