Reference model and building blocks for AI-as-a-service in transaction processing in broker pools (REBAKI)

funded by:

In the joint project “AI-as-a-Service: Automated Process Handling in Broker Pools,” the University of Rostock is working with KeyVance IT GmbH on innovative solutions for automating central business processes in the insurance industry. The goal is to develop a modular AI-supported industry solution that makes the processing of documents and transactions in broker pools significantly more efficient and, at the same time, more transparent.

The individual project REBAKI at the University of Rostock is dedicated to the development of a generally applicable reference architecture that maps typical processes and data structures in broker pools, thus creating a basis for automation. Building on this, building blocks for a learning system are being designed that will enable, among other things, documents to be classified automatically and relevant information to be extracted in the respective context. These building blocks are designed in such a way that they can be flexibly combined and adapted to different requirements using methods such as transfer learning. In addition, a methodology is being developed that allows customer-specific solutions to be systematically designed and provided as a cloud service.

The results of the project not only expand research at the University of Rostock, but are also relevant for teaching in some cases, allowing students to benefit from the latest developments at an early stage. At the same time, the concepts are being tested in close cooperation with industry, creating a close link between science, education, and practice. In this way, REBAKI is making an important contribution to the digitalization of the insurance industry and to strengthening Mecklenburg-Western Pomerania as an IT location.