2 years ago
On a steady course towards the autonomous factory

- There is no doubt that with the help of AI we get better decision support and in time autonomous production plants. But to get there leading suppliers first need to develop scalable and open solutions, says Lars Almkvist, Valmet.

Valmet is the leading global developer and supplier of process technology, automation and services for the pulp, paper and energy industries. Constantly further developing and technologically improving machine equipment, service and customers' processes has always been part of Valmet's DNA.

Construction of new factories involves deliveries from several suppliers. This means that customers therefore demand open and flexible solutions that work together.

- In the end, no one wants twelve different control systems in the plant, but what is demanded is a single homogeneous solution that integrates all the different basic control systems, overall controls with process optimization, simulators and analysis systems that have been delivered and implemented, says Lars Almkvist.

Great focus on development

Valmet annually makes major development investments with many different initiatives on, among other things, industrial internet and IoT with everything from simulators, overall process optimization, to more IoT-specific tools including data analysis and prediction.

- There is huge amounts to be saved for customers by constantly optimizing operation and maintenance, says Lars Almkvist.

Many of these ventures also involve AI. The development projects have resulted in several industrial AI applications, which, among other things, in time help to predict breakdowns and web breaks in paper machines.

- For our customers, it is extremely valuable to see trends and patterns and to be able to predict quality in advance and predict maintenance needs.

Challenges with AI

In pulp and paper, it is very much about analyzing data over long periods of time.

- The availability of data and the quality of data is crucial for a good result. As new sensors are developed and data collection is improved, the possibilities for analysis, prediction and optimized / autonomous operation improve. We are working a lot with this right now, says Lars Almkvist.

The dream of the autonomous factory

An increased use of AI to better understand, optimize and control the processes is a prerequisite for being able to build the factories of the future, believes Lars Almkvist.

- There is no doubt that with the help of AI we will get better decision support and in time autonomous production plants. But to get there it is necessary that leading suppliers strives to develop scalable and open solutions.

- Our customers have all painted the autonomous factory as realistic to reach within 5-10 years. And we are on our way. But to reach this goal, several different factors must be in place. There are many challenges on the road. Pulp production is a complex process.

Lars Almkvist
Director Simulators & Process Control at Valmet


The Autonomous Production