Can IoT be Effective without AI?
There’s no lack of data in our market today…. and many existing systems struggle to filter the signal from the noise. Our goal is not data for the sake of data, but to embrace the cloud and analytical technologies to deliver more expert insights to the right stakeholders at the right time.
Nick Farrant: Rolls-Royce Senior Vice President
Internet of Things (IoT) means the network of connected sensors, devices, vehicles. The parts of this network are connected, typically in a wireless way and are able to communicate with each other, to transfer the data. So the core part of IoT is data. However, data alone does not make sense. You need to analyze data, find patterns and dependencies, apply algorithms to get the sense and value from the data you have. The answer to all that is AI.
Getting the Data from IoT Solutions
IoT provides ways for digitalization of business in many industries. With IoT solutions implemented, a company can get data about various, sometimes unexpected aspects of its business. But if you don’t have any IoT solutions implemented in your company yet, you should think carefully about what solution you need. A good point to start in IoT is to answer the question: what type of data can be helpful for my business goals?
Let’s discuss what benefits you can derive from IoT generated data. I’ve picked up some examples of such benefits just to demonstrate how versatile and non-obvious they may be.
- Enhance service by providing GPS-tracking for in-car delivery
- Improve person’s health and well-being by monitoring their sleep time
- Improve your product quality by estimating damages via a video camera
- Save money by providing intelligent cooling system in the data center
- Improve safety and security using microphones, videos, sensors, and other connected devices strategically placed throughout the area, so that the company security service can readily identify when and where a penetration to the restricted areas is occurring
- Optimize assets by introducing assets/equipment monitoring and proactive maintenance
- Reduce expenses by tracking fleets and monitoring their fuel consumption
After defining the data that can be useful, you can choose the devices and sensors capable of providing this data. Now the list of possible measurements is really huge and limited mostly by the imagination... and price. You can measure temperature, distance, volume, light, sound, position, and much more.
Deriving Benefit from IoT-generated Data with AI
AI can help to deal with data in several ways:
- To make an active exploration extracting the data in an intelligent way. For instance, a smart vacuum cleaner that is exploring a room according to an optimal route designed independently by the device.
- To detect and classify patterns from your data. For example, this can be used for predictive maintenance to forecast when maintenance should be performed and avoid failures and downtime. Another example would be detection of damaged parcels based on the camera video for the delivery service.
- To perform anomaly detection. AI can detect any anomalies in the data and warn an operator.
- Optimization. AI can independently perform direct optimization of the target process. A nice example is a smart city where AI-driven systems manage public lighting to make it cost efficient, control traffic lights to reduce delays and traffic jams.
Examples of AI and IoT Synergie
Examples of AI and IoT used in combination to create smart solutions and get extra value from the data received from IoT sensors are numerous. I’ll name just a few to demonstrate how practical and reasonable AI + IoT solutions are. A good example of cooperation of IoT and AI is Nest smart thermostat solution. Nest device “learns” temperature preferences of it’s users and creates personalized schedules ensuring users’ comfort. At the same time, it minimizes AC usage during rush hours by achieving the preferred temperature in advance.
In addition to producing cars, Rolls-Royce manufactures aircraft engines, which nowadays are loaded with sensors producing terabytes of data. The company decided to make a good use of aircraft engines sensors data: minimize fuel consumption and provide extended reliability and maintenance service. The solution they’ve created analyzes the terabytes of data in the cloud using AI approaches. Data modelling allows predicting when a certain engine may require maintenance, determining which factors influence fuel consumption the most, and more.
And of course, I cannot help but mention autonomous driving vehicles, such as those developed by Tesla Motors and Google. The concept of autonomous driving is built on processing data from all car sensors using machine learning techniques and predicting the behavior of cars and pedestrians in various circumstances. What makes self-driving cars so unique is well described in Elon Musk’s words cited in an article in Fortune: “The whole Tesla fleet operates as a network. When one car learns something, they all learn it. That is beyond what other car companies are doing…”
Conclusion: AI and IoT is a Dream Team
As we can see, IoT + AI combination provides endless opportunities merely for any business in any industry, including logistics, air transportation, ecommerce, retail, municipal services. Continuous growth of processing power, available data-volume and algorithms make these technologies very promising and desirable for investments. Whatever business you have, you get valuable insights, optimize processes, and introduce better planning with IoT and AI working in combination.