AIoT is about advanced algorithms detecting patterns from gigabytes, terabytes and even petabytes of data generated by sensors embedded in physical devices like refrigerators, factory machines and wind turbines. It’s about machines acquiring the knowledge to make decisions with no human intervention. It’s the smart technology that someday may drive your car, handsfree.
While AIoT isn’t the snappiest abbreviation ever coined, the technology behind it is a hot topic of discussion in boardrooms around the world. Why? Because it’s a way for companies, especially in manufacturing and logistics, to extract value from the deluge of data they’re amassing. It allows them to take a more proactive approach to minimizing unplanned downtime, identifying areas for improvement and tapping innovative new revenue streams – all of which improves the bottom line. Gartner predicts that by 2022, more than 80 percent of enterprise IoT projects will include an AI component, up from only 10 percent today.
Other AI technologies such as speech recognition and computer vision can also help extract insight from data that used to require human review. In the eyes of Altmann and other experts, the future of IoT is AI.
Take-up of AIoT is growing. SAS is working with GE Transportation, for instance, to bring IoT and AI to its locomotives in North America. Hundreds of IoT sensors mounted in the locomotives are generating billions of data points per second as they pull railcars as long as 7 kilometers with more than 100 kilotons of cargo. That’s too much data to send and process in the cloud. “The locomotives in the front and back are equipped with an edge solution to perform analytics on board,” Altmann says. “Our analytics software provides real-time information on how to optimize fuel consumption and overall operations.”
Fleet management for commercial vehicles is another area being reinvented through AIoT. Sensors in around 200,000 Volvo trucks in the United States, for instance, are collecting all sorts of data in real-time that is transmitted via telematic devices in the vehicles. The system, which uses SAS analytics technology, enables real-time monitoring of parts and remote diagnostics as well as over-the-air software updates. The company is now applying its AI know-how to that data to offer advanced predictive maintenance capabilities.
Underway in the IoT world for more than a decade, SAS is now expanding its expertise in AI areas such as machine-learning, computer vision and predictive analytics. Coupled with that is the company’s skills in helping enterprises manage the generation, collection, analysis and storage of IoT data. Edge, fog and cloud computing are buzzwords, not to mention streaming analytics. “Some data you want to analyze on the edge of your network, some on the premises (fog) and some up in the cloud,” says Gerhard Altmann, Senior Director of Global Manufacturing Industry Practice at SAS. “And some you want to monitor and analyze on the fly. That’s where streaming analytics comes in.”
In this increasingly connected industrial environment, AI is the “brain” that analyzes and makes decision based on the data compiled by armies of sensor-embedded things. Compared to traditional business intelligence tools, machine learning can make operational predictions up to 20 times earlier and with greater accuracy.
In one of the largest AIoT partnerships to date, SAS has teamed with Siemens to help industrial customers implement the advanced technology. Under the partnership, SAS has embedded its streaming analytics software into Siemens’ IoT operating platform MindSphere, which connects industrial assets, such as machines, plants and entire fleets. The new service provides a comprehensive, open AI framework for industrial IoT.
The partnership goes to the very heart of Industry 4.0. This next-generation smart manufacturing marries physical production and operations with digital technologies such as AI, machine learning, cloud computing and big data to create a more holistic, seamless ecosystem for companies focused on manufacturing and supply chain management. It represents a paradigm shift in industry.