REVAL | Finance & Insurance

Identifying and Prioritizing AI Use Cases

REVAL – Identifying and Prioritizing AI Use Cases
As an expert in financial data, Reval Austria GmbH was looking for methods to identify AI application scenarios and prioritize them based on facts. By means of data value check developed by Know Center, the acquired use cases applied a cost-benefit logic to significantly reduce the risk for subsequent implementation projects.

In times of advancing digitization, one of companies’ main strategic goals is to generate business-relevant added value from data. The important question is: Which data-driven application scenarios can be implemented with the highest possible benefit and the lowest possible risk? It was Reval’s goal to establish a company-internal competence center for AI and machine learning and utilize machine learning for their treasury and risk management software. Using the data value check, a portfolio of systematically collected and evaluated data-driven AI application scenarios was created. This central knowledge base supported management in selecting the optimal AI application scenarios of potential added value and risk. Even prior to the project start the risk of failure was minimalized significantly. Together with Reval, Know-Center’s project team mapped out and documented a total of 18 rough concepts in a participatory manner. The concept’s development was driven by two vital assumptions:

Efficiency: Concepts with poorer chances of success are filtered out at an early stage, if possible. Resources are only used to pursue more promising concepts.

Transparency: All evaluations and decisions are based on comprehensible criteria. All concepts and the associated information are graphically prepared, presented and handed over to the customer.

A profound understanding of the situation and potential solutions were developed. The domain-related input provided by the customer’s employees and the consulting team’s technological expertise served as a basis. The concepts were checked systematically and in several steps by both teams. Less promising concepts were gradually eliminated from the pipeline. Three top concepts finally emerged, illustrating how, why and for whom the concepts created added value and the associated risks on a technical, economical and infrastructural levels.

The participatory approach ensured an organizational buy-in on an operational level, which allowed a smooth transition to the group-internal competence center for AI and machine learning. The top concepts were implemented in follow-up projects while some of them were integrated into the production software.

 

More information:

Data Value Check