Operationalization of Machine Learning Models at VW Financial Services

This case study details the implementation of machine learning (ML) models at Volkswagen Financial Services (VW FS) to enhance fraud prevention in credit risk management. The document outlines the challenges faced due to a high volume of manual checks on earnings certificates, which were time-consuming and resource-intensive. To address this, VW FS developed an ML-enabled statistical forecast model that evaluates the probability of fraud in credit applications. The operationalization of this model was facilitated by the ACTICO Decision Management Platform, which bridged the gap between data scientists and existing IT applications. As a result, the number of manual checks was reduced by 80%, leading to significant efficiency gains and annual savings exceeding 1 million Euros. The case study emphasizes the importance of integrating advanced analytics and automation in financial services, showcasing how the ACTICO platform supports the continuous learning of the model to detect fraud patterns effectively.