Manufacturing

Watson and AI-powered manufacturing dominate 2018

Share this post:

As VP of Offering Management for IBM Watson IoT, I am often asked what Watson can do for manufacturing. As it happens, I had the chance to speak to Ralph Rio, of ARC Advisory Group about exactly this a few weeks ago at the ARC Industry Forum. The answer lies in artificial intelligence (AI). AI-powered solutions enable manufacturers to aggregate data from multiple sources and apply AI for better outcomes. This means healthier assets and equipment, more efficient operations, and reduced costs.

To understand the benefits of AI-powered manufacturing, we start with data and how manufacturers are using it to understand and optimize their assets and operations. The problem here is that many aren’t using it. There’s simply too much plant data to make sense of. As much of it is siloed, its potential remains unrealised. With IBM’s AI-powered solutions, however, we can leverage machine learning and artificial intelligence to integrate vast quantities of plant data – and use the resulting holistic data picture to improve throughput, quality, cost and fulfilment.

Machine learning and artificial intelligence: making sense of the data

Machine learning and artificial intelligence represent a revolution for understanding and acting on plant floor data. With these capabilities at our disposal, we can train models much faster than ever before and better integrate data from disparate sources. For example, while it typically takes three to four weeks to train an operator in quality inspection to achieve 80% accuracy, with AI, this figure shrinks to just hours to achieve 95% accuracy. That’s a considerable time reduction and much higher level of accuracy than was previously possible.

Artificial intelligence and machine learning algorithms like the one in the example above are a game-changer for manufacturers. They allow us to sift through the available data at faster and faster rates, to yield insights that enable us to manage and optimize assets and operations. This means better predictive maintenance, better approaches to quality inspection and increased production efficiency. And the more data there is, the better the insights we can glean from it, and the better informed our decision making can be.

The winners in manufacturing will be the companies with the most data at their disposal, and the right tools to leverage it. Our asset management and optimizations solutions can help you do just that.

The journey forward

In the next installment of this blog, I’ll talk about how to get started with AI-powered manufacturing. In the meantime, take a look at the video for a deeper look at our AI-powered manufacturing solutions. For more information about our portfolio, visit our website.

Vice President, Watson IoT, Offering Management

More Manufacturing stories
By Mark Swinson on February 19, 2019

The 3 Ps of customer engagements: getting started with IoT

When you're ready to make the most of your data through an IoT Platform, you have options on how to get started. This article will help you determine if you need a Proof of Concept (PoC), a pilot or a full production instance of the IoT Platform

Continue reading

By Mark Swinson on February 5, 2019

Rethinking your business with IoT: your product “as a Service”

The more you embed sensors and actuators into physical assets, the more potential you have to change the business models for those physical things. This article will explore the implications of these changes and what they mean in our new "as a Service" world as we progress from purely physical assets to a purely digital ones.

Continue reading

By Jiani Zhang on January 22, 2019

Having data versus using data from your platform

Data is the lifeblood of the Internet of Things. However, there’s a huge difference between having data and using data. That’s why this post explores the value of smarter data. Previously, we talked about the fundamentals of data management. Analytics is the next step in the journey. Not too long ago, IoT meant “connect things […]

Continue reading