Durst looks to AI for machine uptime

Wide format and labels inkjet print systems developer Durst has started a project to develop predictive maintenance and parts replacement on its printers, through artificial intelligence. 

Durst already has intelligent sensors and software evaluation on its printers, and says the move into AI is the next step, and will enable it to predict maintenance and parts updating in advance, according to individual machine situations.

The project is a further component in the Durst vision of a smart factory, where networked infrastructures, intelligent production systems and innovative software enable an automated business process. The printing industry as a whole is heading towards smart factories, with various developers looking to them as the future. If drupa had taken place the smart factory would have been a key theme.

Matt Ashman, CEO of Durst Oceania said, "Durst is actively working towards AI and the smart factory. We already have end-to-end ecosystems and high automation, and the smart factory, with AI playing a significant role, is very much on the horizon. This new project is part of bringing that vision into reality."

Durst says while all its machines have regular maintenance programmes they are all operating under different conditions with different loadings. Christian Casazza, customer service director of the Durst Group says AI will “make these predictions and interventions before an emergency even more efficient, and to be able to apply them even to complex, causal relationships. This is a decisive advantage, especially in times when international traffic is restricted.”

 

Durst is starting the EU-funded project, named Premise, with unibz, a European university, and a manufacturer of snowmaking systems, TechnoAlpin. The project is headed by Johann Gamper (no relation to Durst CEO Christof Gamper), professor and vice-rector for Research at the Faculty of Computer Science. He says, “In the Premise project, we are calculating appropriate algorithms that make predictions about maintenance requirements, including for sub-areas. In this project, we can test technologies that we have been researching for years with our industrial partners on the basis of specific case studies and adapt them to specific requirements. We are thus contributing to technology transfer - an important mission of unibz."

The project will run until July next year, and may be extended after that. Michael Deflorian, business unit manager Software & Solutions of the Durst Group, says, “With the predictive maintenance developed in the project framework, the machine learning techniques used will in future trigger the maintenance of the printing systems independently in order to guarantee predictable and trouble-free operation. "