IBM Watson Thinks Its Way to Factory Floor

New Augmented Intelligence/AI Assistant for Manufacturing solution enables factory inspectors to bring new levels of efficiency to manufacturing processes.


IBM Watson is branching out. Most well known for defeating multiple-time champion Ken Jennings in "Jeopardy" several years ago and for helping doctors determine the best meds for their patients, Watson is now going industrial; IBM is bringing its formidable artificial intelligence to the factory floor.

At a moment when manufacturers are starting to experiment with how to use the internet of things and AI to improve the flow of operations, Watson is stepping in to provide analytics help.

Along these lines, IBM at the Hannover Messe conference April 25 unveiled a new Augmented Intelligence/AI Assistant for Manufacturing, which enables factory inspectors to bring new levels of efficiency to manufacturing processes. Watson's role inside the AI Assistant is to help both managers and factory workers spot and reduce manufacturing defects and increase overall assembly-line efficiency.  

Based on early beta testing of a production cycle that typically takes eight days with a half-day required for needed visual inspection, the new AI Assistant reduced inspection time by 80 percent and cut manufacturing defects by 7 to 10 percent, IBM said.

IBM Joins Forces with Automation Expert ABB
Using its global ecosystem of manufacturing partners and clients, the company is expanding its already significant IT footprint thanks to a new collaboration with Zurich-based ABB, a large global manufacturer of industrial automation and robotics. Working with IBM, ABB is using Watson as its cognitive assistant to help clients tackle some longtime industry challenges, such as improving quality control issues on a factory floor.

ABB's domain knowledge and portfolio of digital solutions along with IBM's expertise in artificial intelligence and machine learning could turn out to be a viable option for manufacturing decision-makers. The first two joint industry solutions powered by ABB Ability and Watson will bring real-time cognitive insights to the factory floor and eventually to smart grids.

The new suite of solutions developed by ABB and IBM and demonstrated at Hannover Messe this week are aimed to help companies address in a completely new way some of their biggest industrial challenges, such as improving quality control, reducing downtime and increasing speed and yield of industrial processes. These packages will move beyond current connected systems that simply gather data, to cognitive industrial machines that use data to understand, sense, reason and take actions supporting industrial workers to help eliminate inefficient processes and redundant tasks, IBM said.

Moving into New Industrial Worlds
"This important collaboration with ABB will take Watson even deeper into industrial applications--from manufacturing, to utilities, to transportation and more," said IBM Chairman, President and CEO Ginni Rometty. "The data generated from industrial companies' products, facilities and systems holds the promise of exponential advances in innovation, efficiency and safety.

"Only with Watson's broad cognitive capabilities and our platform's unique support for industries can this vast new resource be turned into value, with trust. We are eager to work in partnership with ABB on this new industrial era."

For example, ABB and IBM will use Watson's artificial intelligence to help find defects via real-time production images that are captured through an ABB system, and then analyzed using IBM Watson IoT for Manufacturing. Previously these inspections were done manually, which was often a slow and error-prone process.

As parts flow through the manufacturing process, the solution will alert the manufacturer to critical faults not visible to the human eye in the quality of assembly. This enables fast intervention from quality-control experts. Easier identification of defects impacts all goods on the production line, and helps improve a company's competitiveness while helping avoid costly recalls and reputational damage.

Smart Grids Next on the Agenda

In another example, ABB and IBM will apply Watson's capabilities to predict supply patterns in electricity generation and demand from historical and weather data. This is designed to help utilities optimize the operation and maintenance of smart grids, which are facing the increased complexity created by the new balance of conventional as well as renewable power sources.

Forecasts of temperature, sunshine and wind speed will be used to predict consumption demand, which will help utilities determine optimal load management as well as real-time pricing.

According to Business Insider Intelligence, the installed base of manufacturing IoT devices is expected to swell 3 times--from 237 million in 2015 to 923 million in 2020. By that year, manufacturers will spend approximately $267 billion on the IoT.  Manufacturing of these devices require the highest level of inspection for quality during every stage of production.  

More than half of these quality checks involve visual confirmation, which helps ensure that all parts are in the correct location, have the right shape or color or texture, and are free from scratches, holes or foreign particles. Automating these visual quality checks is difficult due to volume and variety of products, in addition to the fact that defects can be any size--from a tiny puncture to a cracked windshield on a vehicle.

ABB is a 130-year-old technology maker of electrification products, robotics and motion, industrial automation and power grids, serving customers in utilities, industry and transport and infrastructure globally. ABB operates in more than 100 countries with about 132,000 employees.

Chris Preimesberger

Chris J. Preimesberger

Chris J. Preimesberger is Editor-in-Chief of eWEEK and responsible for all the publication's coverage. In his 15 years and more than 4,000 articles at eWEEK, he has distinguished himself in reporting...