Services
The client is an American, multinational, investment bank and financial services company. It needed a solution that would consolidate data silos and minimize costs and risks. In addition, it was looking to consolidate solutions across lines of business.
The client was facing issues with existing risk models and had multiple data silos. Its auto, credit card, and mortgage risk models only worked on siloed data sets leading to an incorrect risk score. The client was also unable to implement enterprise-wide predictive analytics.
We, at Mphasis, addressed all of the client’s regulatory concerns. We revamped the retail model scoring solution by creating a model deployment system to break data silos and enable predictive analytics on unified data model. We implemented the following through our solution:
• Developed Component Model Deployment System (CMDS)
• Architected and implemented unified data layer across lines of business (LoBs), thereby providing businesses greater feasibility from unified data asset view
• Developed unified platform across LoBs to reduce deployment time of new risk models (auto/mortgage/credit card loans)
• Implemented security protocols to protect PII data while data is in rest and in motion
• Created multiple data integration channels with traditional database systems, existing Hadoop data lake, and business rule defined excels
Through our efforts, the client optimized the processing time from ~20 days to under a day. It also gained other benefits such as:
• Aggregation of seven years of customer data into a ‘single source of truth’ to enable auto, credit, and mortgage loan models to work on unified data assets
• Implementation of risk models as a service, thereby serving customers quickly
• Improvement of operating margin by reducing risks
• Elimination of software licensing fee