The IoT is built on many innovative technologies: Louis Fourie |
Smart devices and commodity sensors generate an immense amount of data
THE INTERNET of Things (IoT) plays a major role in the realisation of the Fourth Industrial Revolution. But what is this IoT that everybody talks about?
The IoT is an interconnection of computing devices and sensors embedded in everyday objects. These objects – devices, vehicles, machines, smart home appliances, and other items – constantly collect and exchange data via the internet. Currently, about 17 billion smart devices and sensors are connected to the internet and can be remotely monitored and controlled.
As a result of cheap processors and wireless networks, it is possible to turn anything, from a pill to an aeroplane engine, into part of the IoT. This enables devices that would normally be dumb to collect and transmit data without human involvement.
The IoT is built on many innovative technologies, such as application programming interfaces (APIs), which connect devices to the internet and makes the creation of applications and extraction of data possible.
Smart devices and commodity sensors generate an immense amount of data that are analysed in real time and used to make improvements. This is where predictive analytics and Big Data management tools become useful. Artificial intelligence (AI) and machine learning are also used to add context to data and activate actions without any human intervention.
Other keys IoT technologies are the cloud and radio frequency identification (RFID) chips. Cloud-based IoT platforms gather device data, link devices to back-end systems, ensure IoT interoperability, and run IoT applications, while RFID tags – lowpower chips that communicate wirelessly – are used to capture digital data encoded in badges, cards, or smart labels to help track the location of people or equipment.
The first ”smart device“was linked to the internet already in 1982, when a modified Coke machine at Carnegie Mellon University became the first internet-connected appliance, able to report its inventory and the temperature of replenished drinks.
Since the first connected machine, IoT technologies disrupted many traditional business models and established huge opportunities for companies to create new services based on real-time sensor data. It is often used to automate business and manufacturing processes, remotely monitor and control operations, optimise supply chains, and conserve resources. One of the biggest benefits of the IoT is that it allows people and companies access to more data about products and internal systems, and thus a greater ability to make changes as a result.
We are currently in the first generation of IoT, mostly collecting data from dumb devices and sensors, whereafter it is aggregated in centralised cloud solutions. In the coming generations, we will see much more of edge analytics, intelligent networks and IoT devices as actuators.
The IoT is indeed getting smarter. Companies are incorporating artificial intelligence – in particular, machine learning – into their IoT applications to automatically identify patterns and detect anomalies in smart sensor and device data such as temperature, pressure, air quality, humidity, vibration, and sound. It has been proven that machine learning can make operational predictions up to 20 times earlier and with greater accuracy.
AI applications for IoT further enabled companies to avoid costly unplanned downtime of equipment. Predictive maintenance – using machine learning and analytics to identify patterns in the constant streams of machine data – can predict equipment failure ahead of time with a reduction in planning time of 20 to 50 percent, an increased uptime of 10 to 20 percent and a cost-saving of 5 to 10 percent.
Just as machine learning can predict equipment failure, AI-powered IoT can also predict operational efficiency by processing constant streams of data and finding patterns invisible to the human eye. The IoT sensors and machine learning tools of a shipping fleet operator established that cleaning the hulls more often – an expensive, downtime-causing exercise – actually increased the fleet’s overall profitability since smooth hulls enhance fuel efficiency enough to outweigh the increased cleaning costs.
Similarly, predictive IoT are used to discover fraudulent behaviour at bank ATMs, predicting car driver insurance premiums based on driving patterns of the driver, identifying potentially dangerous stress conditions for factory workers, and monitoring law enforcement surveillance data to determine likely crime scenes ahead of time.
The IoT also combines with the power of artificial intelligence, blockchain, and other emerging technologies to create “smart hospitals” of the future. Medical devices, patient monitoring tools, wearables, and other sensors send data to the cloud via the internet.
The data is then analysed and actionable insights for chronic disease management are created. Predictive analytics becomes very important to predict the deterioration of patients with implanted cardiac devices, internet-connected ventilators, imaging systems, vital signs monitors, and anaesthesia machines.
Strangely enough, AI and the IoT are even changing the way we watch golf. The PGA Tour collects 32 000 data points at every event, and they have about 174 million shot attributes in their database. Using advanced analytic tools and machine learning to review all the data, they provide broadcasters with compelling ideas and concepts that they can use to create personalised stories.