Data Re-architecture Consulting
Re-modelling of enterprise data warehouse to achieve flexible BI reporting with reduced turnaround time for large insurance company in Asia Pacific
Challenges
- Data Model :
- Hindered self-service BI reporting.
- Higher TATs in meeting business reporting requirements.
- Insufficient data insights.
- Higher IT support costs.
- Higher time-to-market.
- AI-Analytics | Process Automation :
- Underutilized AI-analytics potential.
- Impacted self-service AI-analytics capability.
- Unexploited avenues of AI-analytics due to absence of business use cases.
- Low operational efficiency due to lack of AI-enabled process automation.
Our Solution
- Effective & flexible data model architecture enabling self-service BI reporting with lower IT maintenance costs.
- Well-defined AI-analytics business use cases.
Final Result
- Data remodeling in stepwise manner yielded the following ensuing faster-time-to-market, faster BI reporting and higher data insights with lesser IT-dependence:
- Data Lake Storage Layer as the primary source.
- Unified STAR Schema Model – for the EDW.
- Summary tables for quick-persistent storage-based reports/dashboards.
- ABTS for persistent storage-based analytics.
- Data Lake [HDFS] as key source for ad-hoc analytics.
- Implementation of Umbrella Business Cases through the following analytical models (creating positive impact on loss ratio, operational costs, enhancing productivity etc.) around AI-analytics within the space of Life and General Insurance:
- An early Claim Prediction Model.
- False Medical Prescription Entries Detection Model with AI-based Workflow.
- An accurate Lead Generation Predictive Model.
- An accurate Claim Abuse/False Claim Predictive Model.
- Customer Churn Prediction Model.
- Unfavorable Risk Outcome Prediction Model.