Robotics Enabled Personal Loan Underwriting Process for a Large Retail Bank
Typically, in Asian banks the volume of new customers and new accounts on both liabilities and assets side are very high. The extent of penetration of the internet, digital literacy among rural and semi-urban terrains, speed of processing are some of the factors which constrains Asian industry to manage a medley of digital and physical application forms for data collection. High volumes coupled with the ever demanding turn around times (TAT) owing to competition and renewed focus on cost reduction paves the way for implementation of new age solutions. For one of our clients, a large private sector bank, we have devised a Robotics Loan Process Solution that has three distinct features, which enables near 100% automation and hence all the associated benefits of cost optimization, improved TAT and lower risk of fraudulent processing of information are passed on to client.
Feature 1: Physical form converted into digital data by using ICR/OCR technology. As ICR/OCR has limited success rate in case of manual script, to enhance accuracy we have introduced what we call a ‘Robotics Process engine’. The extracted data using an IC/OCR engine passes through these automated rules; a ‘robot’ validates the data collected and alters values based on a pre-processed database. The database will be enriched with every correction that is made and has the ability to store and retrieve millions of patterns/combinations. Robotic process automation (RPA) is the application of technology that allows to configure software or a ‘robot’ to capture and interpret existing applications for processing a transaction, manipulating data, triggering responses and communicating with other digital systems.
Feature 2: Writer Information’s Loan Processing Solution has been created with the flexibility of capturing KYC information from multiple document types, namely the Passport, PAN Card, Aadhaar
Card and utility bills along with a digitized copy of the KYC document. The data capture phase acts as the initial inwarding phase of an enrolment process and integrates with a decision engine. The decision engine has been built using analytical models to derive inferences of data captured and cross check the veracity of the data read by the application automatically. Checks and balances ensure there is an additional layer of data validation.
Feature 3: This processed data helps to underwrite the credit score and then pass the transaction to the Core Business Application for credit decision. Credit decision involves a delegation of powers and hence has to be routed to different stakeholders for approval. To facilitate auto-routing of the proposals, a work-flow tool was integrated with decision engine and this makes Writer Information’s solution fully integrated and automated.
The other highlight of the Writer Info-rmation’s Loan Process Solution is that this can work in a ‘tabbased’ Data Capture model for single case handling as well as in a server-model for bulk processes across thousands of records using the robotics solution.