The internet of things and the industrial internet
While it is often claimed nowadays that the internet has changed everything, only a few industries have been truly disrupted by the internet so far. The media industry is perhaps the most obvious example, especially the music business, which has been completely transformed by the ability to access, share and purchase music online. Other industries have been affected to a lesser extent. For example, the marketing and sales side of the travel and the retail sectors have witnessed growing market share by online merchants at the expense of traditional travel agencies and retail outlets. However, most other industry sectors have not been much affected by the internet to date.
While smart phones are now ubiquitous, the proliferation of other smart devices with embedded intelligence, the ability to sense and the ability to communicate with other devices, has only just begun. According to some estimates cited by O' Halloran, there could be as many as 50 billion "smart things" in the world by 2020. Attention is now focused on personal and consumer devices but smart things will extend beyond this to areas such as the transportation and industrial sectors. The "industrial internet of things" or "industrial internet" is a term that refers to the seamless integration of the physical and digital worlds through networked sensors, actuators and software. It draws together fields such as machine learning, big data, the internet of things, and machine-to-machine communication to ingest data from physical objects, analyze it (often in real-time), and use it to control and adjust operations.
A distinctive aspect of the industrial internet of things is the focus on industrial and commercial systems such as factory machinery, jet engines and MRI scanners by adding intelligence to those things using a set of actuators, sensors and embedded hardware. From a technology point of view, the rise of industrial internet of things was driven by four trends:
The low cost of computing power and sensors allows them to be embedded into everyday devices and equipment economically;
The availability of ubiquitous data network which allows equipment to be monitored, tuned and acted on much faster;
Fast and relatively inexpensive advanced analytics that allow organizations to harness the massive amount of data generated by the intelligent devices;
The availability of external data (eg weather, traffic data) that can be used to enrich sensor data to make intelligent decisions.
Tracking: Real-time tracking and monitoring of resources such as where they are, their status and overall usage. This capability can be used, for example, to improve worker safety by monitoring their location and ensuring they work within the perimeter they are allowed. In supply chains, it can be used to track where a piece of inventory currently is located and when it will arrive at its destination.
Optimisation: Optimising the use of resources based on the monitored activities which may involve re-planning or changes to business processes at the system level. For example, if a bus is stuck in a traffic jam, other buses can be re-routed to minimize further congestion. This information can then be shared with other drivers.
Prediction: Predicting the overall "health" of the resource and how soon it needs to be serviced based on historical usage and current condition. In transportation, it can be used to predict when the engine of a bus needs to be serviced. In supply chains, it can be used to predict how much stock is needed based on current location of incoming shipments and upcoming weather conditions.
Actuation is the system's ability to automatically act upon, change or control the physical world. Actuation may take the form of direct controls, as in the case of robots, autonomous vehicles, and automated dispensing of medication. It can also take the form of indirect controls, such as automatic ordering of a part when the system detects the need for a replacement, such as when a tyre needs to be replaced.
Street lighting: Smart lights will only turn on when needed (sensors will detect people in close proximity) and will notify operators when there is a problem in need of maintenance.
Air quality management: Special equipment will be able to measure and predict air quality and then offer incentives to consumers to take mitigating action. For example, consumers could be offered free use of bicycles or larger discounts on their electricity bills if turn off their air conditioners at certain hours. Taking this a step further, if a smart device is installed in the home, it could be programmed to switch off the air conditioner automatically when notified of the discount offer by the air quality device. This is just one example of smart devices talking to each other and acting autonomously. When you combine multiple similar devices and actions, the potential is far greater.
General Electric has predicted that these kinds of actions would increase efficiency and could boost productivity by 1-1.15 percentage points and thereby add as much as US$12 trillion to global GDP. Some are even more optimistic about the potential benefits.But besides benefits, there will also be disruption. Besides disrupting individual industries, the industrial internet will also blur traditional boundaries between industries. In addition, while it is clear that there is potential to improve efficiency, many, especially governments, are concerned about the impact on jobs. With increasingly autonomous machines, wouldn't this eliminate a lot of jobs, especially menial, blue collar jobs? However, according to O' Halloran, many new jobs will be created, including new and better blue collar jobs, for example, technicians and maintenance workers. These jobs could be filled by re-training existing blue-collar workers, thereby addressing the employment problem.
When will we see the unleashing of the potential of the industrial internet? The answer depends on who you ask. Those in the tech industry, who understand the technology and the enormous potential, expect to see results within a year. However, those in traditional industry don't expect much to happen for another decade or so. Perhaps the answer is somewhere in between for while technologies are available, there are regulatory and practical issues to resolve before things can take off. For example, issues of standardization (the mechanisms and protocols of how devices "talk" to each other) and security (how to regulate the ownership and use of data) are just two of the difficulties to resolve. Privacy and security are potential inhibitors of the industrial internet, especially given the difficulty of getting any kind of international consensus on these issues.