Do you have the needed data around yield, throughput, and quality to make sound operational decisions?

Share this post:

Here is a bold idea – your machines, your connected systems, your factories all diagnosing themselves and predicting needs before they affect production. What do you think when you read that? Are you thinking early and accurate identification of defects, improved production, optimized throughput, less waste and no product recalls? You should.

Because that is what IoT and AI, infused in your factory operations, can help you achieve. No matter what your product is, high quality and high yield should always be expected. And for that to happen, it’s time to step away from the traditional manual methods of quality inspection. These are often expensive, error prone and time consuming, and they eat into your capability of keeping up with the demands for fast and default free products.

Do you have this kind of data and knowledge?

Let’s explore further how IoT and AI can help you better monitor operations and production and get a better yield. You can create data on your machines in many ways, including without even touching the machines themselves. For example, using AI-powered visual and acoustic analytics, manufacturers can intelligently create new data patterns and identify defects based on images or sounds. Watch what your machines are doing and listen to them tell you if its healthy.

At IBM we have a solution called IBM Production Quality Insights. This powerful set of tools brings enables monitoring, inspection and insights during the manufacturing, assembly and maintenance process. With the aid of AI, it enables manufacturers to identify the root cause of defects from wide-ranging inspection data, whether structured (process data), visual data (image data) or acoustic data.

This solution equips you and your inspectors with the ability to identify quality problems earlier and to know what are the potential triggers of poor quality. This opens a new world of opportunities where you can predict any constraints to productivity and prescribe remedies to fix issues accurately.  And your inspectors do not need to go through a long training process anymore. They can now have the knowledge of a lifetime experience and more, at their fingertips. At the end of the day, connected organizations are experiencing improved inspection time, without dealing with manufacturing defects anymore.

It’s time to leverage AI capabilities

We have been engaging with countless customers across all types of manufacturing to bring our AI-powered approach to the factory floor. Successful use cases include aluminum smelting energy efficiency improvements, injection molding downtime reductions, steel process quality improvements, increasing uptime at an automotive body shop, and optimizing quality and energy consumption at a cement mill.

All these organizations have leveraged AI capabilities and successfully improved the accuracy of their inspections. They’ve done this whilst reducing inspections costs by up to 10 percent, reducing scrap by up to 3 percent and achieving a higher yield.

Recently, we worked with Shenzhen China Star Optoelectronics Technology Co., Ltd., a large global LCD manufacturer, to deploy the Production Quality Insights solution in their production line. It took only eight weeks to deploy and train the inspectors. And the inspection time reductions are even more impressive: while it used to take 30 minutes or more to inspect each batch of LCD panels, we were able to reduce that to just a few minutes per batch. That’s a massive time saving. It turns out that a picture is worth a thousand man-hours, not to mention less physical strain and comparable accuracy.  The overall manufacturing cycle has been reduced, as well. In LCD production a few hours in cycle time may not sound like much, but imagine what an impact this makes to the business as a whole. Today, every month over a million screens are inspected in the factory and those hours add up quickly. 

How to get started

The demands for improved quality and throughput are getting higher and higher. But that shouldn’t be a blocker in your enterprise’s success. Rather, it should be considered an opportunity to work smarter. I invite you to reach out and find out more about how this solution can improve operations, maximize yield and throughput, and better understand how you can optimize production in your shop floor. Quality matters. Don’t lose it. Learn more about IBM Production Quality Insights today.

More Manufacturing stories

Robots and Node-RED: Rapanui, the eco-friendly clothing factory

Written by Dr Lucy Rogers | April 5, 2017 | Blog, Factories

Rapanui, an award winning fashion brand uses Node-RED, MQTT and Raspberry Pi computers to make ethical and eco-friendly bespoke products at high street prices – no guilt tax involved. One of the services that Rapanui Clothing Limited provides is T-shirt printing. They use open source technology to help reduce the batch size of their T-shirt more

Transforming engineering and product development with IoT

Written by Zach Jory | March 18, 2017 | Automotive, Buildings, Factories

As the Internet of Things becomes pervasive, engineering organizations are exploring the possibilities of IoT and figuring out how to implement in a way that drives tremendous business impact. Products have moved beyond mechanical and electrical components to include complex combinations of hardware, sensors, data storage, microprocessors and software—as well as ubiquitous connectivity. While this more

What is industry 4.0?

Written by Jen Clark | October 12, 2016 | Automotive, Blog, Factories...

Industry 4.0: IoT does smart manufacturing Industry 4.0 manufacturing is all about the increasing connectivity of smaller and smaller components. It uses the data collected by physical things, either built products out in the real world, or components within a machine on the production line, to impact how these things themselves are manufactured. The more more