Story image

How artificial intelligence can prevent machine failures

04 Jul 17

Despite rapid advances in technology, we’re still seeing many stories of product recalls, factory failures and other types of maintenance accidents repeated, year after year.

Just last March, Ford recalled almost 5,000 cars locally after seven burst into flames and more than 50 caught fire overseas.

While mechanical and technical failures sometimes can’t be prevented, most of the time they can be avoided by being spotted earlier enabling recall before a user-facing incident occurs.

Today, next-gen analytics and automation technologies exist that can help manufacturers avoid the burden – and bad publicity – of having to spend millions of dollars dealing with maintenance failure crisis.

Predictive maintenance on the rise

According to reports from McKinsey, predictive maintenance could save global businesses an incredible $630 billion a year by 2025. No wonder why more and more manufacturers around the world are investigating the potential outcomes of investing in this area.

Analytics technologies, which were very inaccessible and complicated to use a few years ago, are now becoming more accessible.

Developments in Artificial Intelligence (AI), machine learning, the Internet of Things (IoT) and data science mean that humans can now rely on technology to spot equipment failures and danger before they occur.

Predicting equipment failures based on smart analytics techniques can result in introducing timely maintenance which can lead to reducing operation costs and keeping downtime at minimum.

Also, being able to properly analyse data from the entire manufacturing chain – including data on how the manufactured product is being used – can reveal meaningful connections and trends. It can not only save time and money, but also streamline the entire production process and boost efficiency at every level.

How it all works

Having AI-enabled capabilities means that advanced machines in factories – and final products used by consumers, i.e. cars - can collect and monitor various types of data 24/7, through in-built sensors.

This data can then be interpreted, in combination with additional contextual information, with the help of data science.

By combining data provided by sensors, contextual information coming from within and outside the business, with predictive techniques, manufacturers can capture real-time status of parts and functions, as well as learn from several scenarios’ history to predict and prevent failures.

To leverage the power of predictive maintenance, manufacturers need to harness the massive amount of data generated by the Industrial Internet of Things (IIoT), analyse it at scale, and translate it into actionable results.

Today, we find that most manufacturers are equipped with IoT sensors and analytics capabilities.

They know how to gather data, and store it, but making sense of it and using it to feed predictive scenarios and automate maintenance processes is still a challenge. To make it work, they need to transition from cognitive adoption to cognitive application.

Leveraging the power of cognitive

Cognitive capabilities involve self-learning systems that use data analytics, pattern recognition and natural language processing to mimic how the human brain thinks and works.

Using cognitive platforms enables data gathered from traditional analytics software to be used repeatedly and for future predictive context rather than just for one off process.

From an application standpoint, a cognitive-first development strategy is required, so every piece of connected equipment can feed into the machine learning engine, taking and merging existing and ‘learned’ datasets with traditional analytics.

Today, platforms exist that can detect random and unknown failures using a combination of unsupervised and semi-supervised learning techniques.

For example, using techniques such as Meta Learning, meaning that a platform will learn and adapt based on experience, can increase the quality, accuracy and timeliness of equipment failure predictions.

Predictive maintenance isn’t just about monitoring machine maintenance on factory floors. It will help manufacturers in a variety of industries to reduce accidents, lower maintenance costs, eliminate unplanned outages and minimise downtime, use less energy, and maximise yield.

Manufacturers investing in the right data science and cognitive technologies today will ultimately be the ones able to unleash all the potential of AI and industrial IoT, which will help them remain competitive, and create new opportunities for revenue.

Article by Mark Armstrong, vice president and managing director of International Operations for EMEA and APJ, Progress

Opinion: Critical data centre operations is just like F1
Schneider's David Gentry believes critical data centre operations share many parallels to a formula 1 race car team.
MulteFire announces industrial IoT network specification
The specification aims to deliver robust wireless network capabilities for Industrial IoT and enterprises.
Google Cloud, Palo Alto Networks extend partnership
Google Cloud and Palo Alto Networks have extended their partnership to include more security features and customer support for all major public clouds.
DigiCert conquers Google's distrust of Symantec certs
“This could have been an extremely disruptive event to online commerce," comments DigiCert CEO John Merrill. 
Schneider Electric's bets for the 2019 data centre industry
From IT and telco merging to the renaissance of liquid cooling, here are the company's top predictions for the year ahead.
China to usurp Europe in becoming AI research world leader
A new study has found China is outpacing Europe and the US in terms of AI research output and growth.
Google says ‘circular economy’ needed for data centres
Google's Sustainability Officer believes major changes are critical in data centres to emulate the cyclical life of nature.
52mil users affected by Google+’s second data breach
Google+ APIs will be shut down within the next 90 days, and the consumer platform will be disabled in April 2019 instead of August 2019 as originally planned.