The Corona crisis has shown that external factors beyond the control of healthcare facilities can rapidly lead to critical shortages in the supply of medical protective equipment such as respirators or disinfectants. These factors include global trade restrictions, long delivery times and low-quality products. There is a missing overarching ecosystem that informs suppliers and manufacturers at an early stage about impending bottleneck situations and adapts supply and value chains in a forward-looking way.
Under the auspices of the competence platform Artificial Intelligence North Rhine-Westphalia KI.NRW, the research project corona.KEX.net was therefore launched. The aim of the project is to develop an intelligent, AI-based early detection system to prevent supply bottlenecks for medical protective equipment.
In addition to the consortium leader KEX Knowledge Exchange AG, various experts from business and science are collaborating on the development of the early warning system. Project partners include KEX-INC Invention Center-Vipro, Fraunhofer IIS/EAS, Fraunhofer IAIS, SBN Data Technologies, Technovation and the Institute Industrial IT (inIT) at the Ostwestfalen-Lippe University of Applied Sciences and Arts.
The inIT is represented in the project by the Diskrete Systeme working group in the field of industrial image processing. "Our research tasks relate to the automatic detection of authentic certificates of medical products. Manual inspection of the certificates is very time-consuming and not manageable for the users. Artificial intelligence can help here," explains project leader and inIT institute director Prof. Dr. Volker Lohweg.
The early detection system and holistic communication and information flows along the supply chain will enable the production and supply chain to respond to changes in medical care facilities at short notice and thus be resilient to strong market fluctuations. Using an algorithm, the system can analyze orders in real time and sound the alarm when certain items are in particularly high volume demand. Suppliers can then be notified early enough to respond.