The Internet of Things and Industry 4.0


This is my second internet of things blog post for my final semester class, emerging trends in technology. In my first blog post I went over the history, future, and current state of The Internet of Things and briefly explained some security risks that have come along with it. If you haven’t read my first post, please do here as it will make this one a lot easier to understand. In this post I want to focus more on how IoT is helping push industry 4.0 into fruition which is a topic I briefly covered in my first post. I will also cover some of the controversies surrounding these new IoT technologies and go into more technical detail regarding some of the infrastructure.

Industry 4.0

As stated in my first blog, industry 4.0 or the fourth industrial revolution, is the idea of manufacturing with little to no human interaction by using technologies like IoT, robotics, and machine learning. Here is a short video that explains it quite well.

The key takeaway from this video is that industry 4.0 exists to produce efficiency and innovation. I chose to include this video over some others because I like how it emphasizes that there are some human aspects that machines will likely never be able to replace such as innovation and creativity. That being said, IoT has shown signs that fully automated factories could be possible in the near future as more IoT applications are being developed and adopted by big corporations. Industry 4.0 would not be possible without the Internet of things as almost every aspect of Industry 4.0 relies on internet connected devices communicating and sharing data with each other. Smart manufacturing is being applied to factories in several ways to reduce human interaction. Three of the biggest applications are Supply chain management/optimization, Predictive maintenance, and asset tracking/optimization.

Supply chain management/optimization

IoT helps optimize the supply chain by using real-time location tracking, condition monitoring, warehouse product location, and contingency planning. As I’m sure you know, location tracking has been in place for many years by several shipping distributors such as Amazon, UPS, and Canada post, but this technology still has room for improvement. Using real-time tracking and contingency planning methods gives drivers the best routes for delivery and lets managers monitor deliveries down the exact second. Automated delivery vehicles are also emerging to optimize supply chains like the ones seen in this video.

Warehouse product location has been used to improve efficiency in some Amazon warehouses by using robots that can locate shelves containing the desired product and moving them to an assembly line and vice versa.

Amazon warehouses typically deploy 400–500 of these machines to drive themselves to stationary employees rather than having their employees walk 10–20 miles each day. Amazon warehouses in the past would have employees walking down aisles with carts all day placing products on empty shelves and scanning barcodes to mark their locations. This method was extremely inefficient as the average of picked items used to be around 100/day but is now around 400/day with the use of these robots.

Predictive maintenance and analytics

Many manufacturers are using predictive maintenance to avoid unnecessary and costly maintenance routines. Using IoT devices machines are able to use performance data to predict when maintenance is necessary and schedule a maintenance routine. Studies show that 82% of machine failure happens randomly while the other 18% happens due to age and wear, proving that time-based maintenance is not cost effective. The predictive maintenance process can be quite complex but I will try my best to break it down into its simplest form.

To implement this technology, we need to identify a few key variables that determine the health of the machine. Using a battery as an example we could use temperature, voltage, and discharge as our variables and then attach sensors to track the output of these variables. Before the gathered data is sent to data storage it has to pass through a few preprocessing steps such as a field gateway and cloud gateway. A field gateway is a physical device that filters the data and is connected to the cloud gateway that ensures safe transmission of data. Once this data reaches the cloud it goes through streaming data processing which ensures a continuous flow of data is being sent to our storage. Now there are 2 different types of data storages that it will be sent to. The first is a data lake which temporarily stores all of the data we gathered from the battery, including data containing anomalies. These anomalies are then sorted out and the clean data is sent to a big data warehouse that contextualizes and timestamps our batteries temperature, voltage, and discharge readings. Finally, the data is ready to be analyzed by a machine-learning algorithm to find patterns and abnormalities in data sets. At this stage the system can predict if the battery is almost at the point of failure and can begin a maintenance process to fix/recharge it and then this whole process would just continuously repeat itself. In reference to the image above, user applications and control applications would not be applicable for the battery example as these would just be used for users to send signals for machines to perform other processes.

Asset tracking/optimization

Asset tracking uses devices to monitor the performance of company assets whether they are machines, vehicles, tools, or even people. This is being implemented to cut out the need for human supervisors to ensure everything is working effectively. Another form of asset tracking that is on the rise is employee monitoring. Amazon web services “panorama” is an IoT surveillance system that uses machine learning to track employees’ movements in excruciating detail to see how efficiently they are working.

Fender, a guitar manufacturing company, has adopted this surveillance technology so they can monitor the length of queues in assembly lines, find flaws and increase productivity. However, technologies like this have stirred up a lot of controversy as many employees feel it forces them into poor or unfair working conditions. There is also an argument regarding the use of this technology to track older employees or employees with disabilities whose data would show that they cannot assemble products as fast as young able-bodied employees. I personally think that this technology can be used in many good ways to improve working conditions for employees by innovating new ways to avoid human error rather than penalizing employees for not being efficient enough.




Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store