By now, you’ve surely heard all about the Internet of Things (IoT), and how it will change the way we interact with the world around us. You’ve also probably heard about the new world of cognitive computing, as exemplified by IBM’s Watson technology. But are you truly ready for the ways that these two things are about to combine to revolutionize the way we think about technology?
First, a few data points and predictions (as stated by IDC’s Vernon Turner):
- In 2015 9 billion devices were connected to the Internet. This includes RFID and machine-to-machine tags, along with consumer devices. This will increase to 30 billion by 2020.
- Currently 4,800 new devices are being connected to the Internet every minute. This will increase to 152,200 by 2025.
- Connected devices will generate 44 zetabytes of data by 2020, and 180 by 2025.
With that much data being created, automation and systems management become much more important. But it’s not enough to just think about data. In fact, an estimated 90% of all IoT data is never acted upon; almost 2/3 of that data loses its value in under 1 second.
Going forward, it will become vital to think about business value of that IoT data–not just numbers and devices. Data becomes more valuable as it’s used, and ROI is proved as that data is actually put into use and leveraged, and incorporated into other IoT data from ecosystem partners. The way to add value is to add a cognitive layer.
Many IoT devices currently function as part of a closed or single-vendor ecosystem. But what happens as they have to start working with devices from other vendors, or as part of a heterogeneous industrial environment? These devices need to be able to think for themselves so things like driver and firmware updates don’t blow up the system. Taking that idea a step further, connected devices need to be able to teach other devices in the system and make them smarter, or help them make better decisions. Cognitive systems make this possible.
In a session today at IBM InterConnect, there were a few examples shown of how IoT data and cognitive computing are starting to work together. The first was from Honda, which recently teamed up with the McLaren Formula 1 racing team. Together, they’re using IBM Watson tools to improve the performance of their cars. Engine data is captured from 160 sensors, monitoring things including temperature, pressure, and power levels. Watson uses this data to optimize power delivery, as well as a new regenerative energy capture system–something new to F1. In this system, heat from the brakes and exhaust is turned into energy, which is saved in a batter (similar to consumer hybrid). The racing team can use recommendations from Watson and communicate them to the driver, suggesting optimal times to pass other drivers, or use the stored energy for an added speed boost.
Watson can also make recommendations for adjustments during pit stops. This data is so useful that the Honda/McLaren team has reduced the duration of the average pit stop to under two seconds. In comparison, the average pit stop was one minute in the 1960s, when the driver was the sole source of data. The next step will be for Honda to apply what they’ve learned to their consumer cars, which may possibly collect data and use cognitive systems to analyze it and recommend repairs before problems can occur.
In a completely different scenario, the city of Shanghai, China is using IoT data to help improve air quality. In a partnership with IBM, they are combining data from 500 different kinds of sensors with weather and satellite data, then feeding that to Watson. The team is using the output to make long-term urban planning suggestions meant to minimize congestion and smog, as well as suggestions that can be acted upon immediately, such as traffic flow restrictions. The hope is that by combining the IoT data with Watson’s cognitive layer, they can make decisions that will help improve public health.
The combination of IoT and cognitive computing holds a lot of promise, and we’re just at the start of what looks to be a sea change. How is your company planning to combine the two? Let us know in the comments!
Latest posts by Matthew West (see all)
- Rocket.Build: Around the World in 15 Days - March 16, 2018
- Live from IBM World of Watson: don’t be afraid of the AI boogeyman - October 27, 2016
- Live from IBM World of Watson: could a cognitive system have written this for me? - October 27, 2016