The use of AI in industrial applications takes manufacturing to completely new levels

It has been said for more than 20 years that advanced analytics would change manufacturing industry fundamentally and solve several of their greatest challenges. Maybe it is time for that old prediction to finally starting to come true.

“Digitalisation has now reached us users, the technology and applications are there and we are ready to take the next step and build it into the organization’s culture.” says Veronika Grozdanovska, CIO at Novotek.

The digitalization of our community has penetrated our society for the past 15 years, at the same time the industry has going much slower in its adaptation. The reasons for that is not always that the technology wasn’t there but that the organizations many times weren’t ready. The tools and infrastructure that digitalization created has now finally also reached manufacturing, and in the last 5 years, a slow acceptance has been created for this new technology. 

Novotek has delivered many types of AI to improve processes and efficiency to our customers. The latest addition is advanced optimization of production flows in real-time.

We have for a one of the largest industrials in Sweden delivered a system optimizing their complex production based on a number of parameters and conditions, and at the same time take into account the actual situation in the plant using machine status, raw material levels and order information. It has given our customer completely new challenges when we have been able to take away all the bottle necks they had inside the factory, and they have now started to use the solutions and optimize  flows outside the plant. It can look like magic, like the customers put it, but with the correct data and access to the right tools it can be done.    

The customers started their journey in analytics by presenting us with an unsolvable task to optimize the crucial bottleneck resource they knew that they had for a long time. Applying classical production scheduling in order to improve the situation had been tested but with limited results. Already early on during the implementation, we could show that AI changed the situation and improve the results significantly. Just weeks into applying the technology, they were able to dismiss the former bottleneck and move on to the next.  

The optimization is now done on a more than ten concurrent parameters and the system creates an optimal production plan that meets not just delivery time and personnel constraints but also a plan to minimize energy consumption, raw material usage and storing requirement. The entire weighing between these different parameters is done via a flexible and simple interface that makes ut easy for the customers to do it on their own.   

On top of this, it may be worth mentioning that the need for flexibility when it comes to how you want to deploy this type of solution is increasing and we see it as completely natural that you should always be able to choose the deployment structure that you see fit. No matter if you want to run your favorite cloud service or deploy locally, we have the right technology to make it work. The ability to run same technology in several levels and to be able to facilitate the needs for accessibility, scalability and latency, is central.

Analytics like AI and ML is something that everyone in the industrial world should include into their plans and strategies for continuous improvements at all levels and areas.