Industrial IoT will give us self-repairing machines

Published by Telefónica IoT Team IoT General, Industry

One of the most promising and direct applications of the IoT is probably the "construction" of an appropriate environment for Industry 4.0. Thanks to connectivity and monitoring, this could help make more efficient industry a reality, industry that is capable of anticipating problems and solving them before they even occur. How will the Internet of Things help us do this?

IoT and Industry 4.0

As we said, the IoT is one of the cornerstones of Industry 4.0. Why? The answer is quite simple: everything is due to real connectivity. The fourth Industrial Revolution, also called smart industry, involves a profound change in the organization of production. This organization will rely more on the smart automation of processes, and we automatically require connectivity in order to coordinate and move this whole system.

Through the network offered by the Internet of Things, it is possible to interconnect assembly lines, monitoring systems, and control or emergency systems, for example. The connected sensors provide a vast amount of information that will be collected through analysis techniques and artificial intelligence, resulting in automatic control for industrial processes, which translates into efficiency, quality and cost reduction.

In 2016, spending by the industry in IoT was estimated to reach 737 billion dollars. The investment will grow to 1.29 trillion dollars in 2020, according to experts from the sector, which augurs interesting growth. But how will this investment be reflected in industrial processes? Predictive maintenance and the industrial application of self-maintenance will be two of the main focal points.

Predictive maintenance: Anticipating breakdowns

As we improve our knowledge of machine learning and artificial intelligence, we are able to detect more complex events, including machine failures due to wear or tear. With specialized sensors, the machinery is able to predict the depletion of a material or the substantial loss of a part’s integrity. In this way, a connected sensor can alert technicians that something is close to breaking down so they can take the appropriate measures as quickly as possible.

In this way, machinery management can be automated, and an operating protocol can even be organised if a failure is expected, so accidents or downtime in the production chain can be avoided and the required efficiency can be maintained at all times. This also saves on repair costs, resources and time. Self-repair and self-maintenance may even be automatically organised one day.

IoT and machines that self-repair

If we can program a production line to detect its faults, we can also programme it to keep running smoothly. This is the premise underpinning the SelSus project at the Fraunhofer Institute for Manufacturing Engineering and Automation IPA. They are developing a technology capable of preventing breakdowns and downtime in their machinery so that production can be organized automatically. The possibility of self-repair is also part of that project.

To do this, they are working on total machinery monitoring techniques. Embedded sensors and other devices connected to the IoT are essential in doing this. With them in place, if a sensor fails, for example, the machinery is able to operate in a "safe mode" while waiting for the repair (which can be automatic, through a line of support robots, or managed by a human technician). As the firm’s experts indicate, Bayesian networks are the key point in developing their self-repairing lines.

Using this type of mathematical model, the likelihood of certain types of events can be calculated. Thus, projects such as SelSus not only rely on sensors but also use predictive mathematical models that complete the forecasts. Nonetheless, creating a machine capable of self-repair still has a series of drawbacks that need to be solved.

One of the biggest challenges is to analyze the massive amount of data coming from all the sources that connect the IoT. Fortunately, Big Data analysis techniques are advancing by leaps and bounds. The ability to work with a huge amount of data, together with artificial intelligence and the ability to monitor the entire production line in real time, are the three backbones of Industry 4.0, a concept in which machines repair themselves without anyone’s help thanks to the IoT.


Telefónica IoT Team