Industrial Artificial Intelligence
Introduction
Sweden has a long and successful tradition of building sophisticated industrial systems, and in the last decade, the deployment of AI-based solutions for data-driven production has accelerated. To stay competitive in the long run, smarter and more advanced AI products and production systems are needed, both as integrated parts of new products and as applied to production systems to optimize these. Cross-industry studies however show that on average, less than half of an organization’s structured data is actively used in making decisions, less than 1% of its unstructured data is used at all, and 80% of analysts’ time is spent simply discovering and preparing data. Notwithstanding the widespread success of AI in the last decade, there are still fundamental challenges remaining in the industrial deployment of AI technology related to performance, accuracy and scalability.
Need
Research in applied AI at MDH focuses on the practical usability of computational methods for analysis and design of real-world problems, including production systems, with the core in approaches from the AI field, such as search, optimization and machine learning. Several groups at MDH are targeting applied AI in different and complementary ways, and the area is one of the most recent and expanding research directions, affecting all three participating research specializations at MDH and forming the core of many INDTECH PhD projects.
Topic
In the area, MDH focuses on analysis and optimization of real-world complex problems in analysis of cyber-physical systems, verification and validation, condition-based maintenance, energy production and transportation and logistics. The direction has been elevated in importance as part of the new Applied AI B.Sc. programme starting in 2020, and the re-focus of the M.Sc.25 program in production and product design towards Industry 4.0 and Applied AI. In the future, simulation and optimization will be used for development of both products and production systems already at a very early stage of the production development. Moreover, intelligent industrial applications will contain AI- based components that work side-by-side with other functional components. Gaining confidence in the safety of such heterogeneous systems is a significant challenge that will be tackled in INDTECH. This together with tools for quick configuration of deployed systems are key elements of Industry 4.0 and strongly addressed here. Industrial partners include major companies, such as ABB, Alfa Laval, Bombardier, Ericsson, Scania and Volvo, as well as many smaller businesses.