Updated: Jun 1, 2020
Welcome back to the unlikely techie blog – great to have you! Let's look at another industrial automation buzzword you might have encountered before: smart factory. Smart factories rely on highly computer-integrated manufacturing processes, high levels of adaptability, and technologies such as artificial intelligence (AI), analytics derived from big data, and robotics. The idea here is to have a factory where everything is interconnected and, therefore, not only can run itself but also self-corrects.
Where Can I Find A Smart Factory?
If you are looking for cyber-physical systems that merge reality and the virtual world, you will eventually find smart factory approaches. This also answers the question. A smart factory to the extend of being fully self-automated and self-maintaining, does not yet exist. The reason for this has a lot to do with its characteristics. A smart factory is defined by visibility, connectivity, and autonomy. While factories can look back on a long tradition of continuous automation processes (the ongoing automation process for me begins with the introduction of the assembly line by Ford in 1913), it's the extent to which smart factories will be able to run without much human intervention. This will (hopefully?) enable them to be more flexible, adaptable to change, and, therefore, more efficient than current factories.
This image shows the capabilities of the Industrial Internet of Things (IIoT), interconnecting industrial applications/networks.
Productivity Thanks To IoT Sensors
All of this can only be achieved by a vast number of IoT sensors, devices, and interconnected networks. Essentially, the challenge here is to have a definite system knowing (we are entering the world of deep learning here) how to read, process, and analyze the vast amount of data generated by these sensors. The capability to successfully operate and fulfilling routine and mundane tasks error-free is going to be increasingly done by robots (this is also where machine learning comes into play), leaving handling exceptions or higher-level decisions to humans.
Benefits and Challenges of Smart Factories
The other benefit of a smart factory is increased productivity in supply chain management - the productivity increases by connecting the plant to other entities in the network (thanks to IIoT). On a larger scale meaning, that all factories of a company could potentially be interconnected, exchanging data on stats such as capacities, maintenance issues, worksheets, schedules, and so on. This sounds great at first glance but also lets my alarm bells go off. Let me tell you why: who has access to all of this data? And, more importantly, who decides based on this data?
There are, however, already implementations today that go into the direction of smart factory building, such as digital twins (which I have covered before), virtual commissioning, and investments into human-machine interaction, also referred to as human-machine interfaces (HMI).
When do you think the first smart factory will start production? Let me know in the comments, or let's further debate on Twitter! Talk to you soon, and until then: stay curious!
As always, here are some helpful links and further sources, some of which I have used to cover this topic:
An article called "Why Everyone Must Get Ready For The 4th Industrial Revolution" by Bernard Marr (it also features an informative video with Prof. Klaus Schwab).
To find out why it is extremely difficult to find success with the smart factory concept, please check out this Forbes article.