Header image  
   
  
 
 
 
 

 
 
BRMS Case Study

An underwriter of home mortgages wishes to automate their underwriting process. The current, manual process averages five business days to complete product eligibility, pricing, and underwriting risk evaluation. Processing volume is restricted by the size of competently trained staff. Points within the business process require the re-key of critical customer information which is subject to quality control issues.

An automated underwriting process is proposed. Specific goals are to reduce processing cycle-times, increase potential processing volume without an increase in staff, and improve consistency and quality.The automated service must not only support the processing requirements of the internal business clients, but must also support the business partners who utilize the same underwriting processes. Implemented as a web service, the new automated process may be executed from a custom web front-end process or through a variety proprietary processes utilized by business partners. Required business documents are exchanged in XML as input to the web service and output is provided both in XML and PDF formats to accommodate both fully automated clients as well as partially automated clients requiring an output suitable for direct customer visualization. Protection of personal customer information is critical throughout the process.

The business rules encompassing the loan underwriting process are proprietary to the customer's business model and are implement as sets of rules within a Business Rule Management System. In this case a proprietary business rule management system chosen by the customer is utilized.

On delivery, the automated process reduces the underwriting process from 5 business days to approximately four minutes. The number of simultaneous processes is only restricted by the amount of hardware the customer wishes to dedicate to supporting the process. Internal processing steps which previously required the re-key of information are eliminated. Business rules are applied from a single image of the rules, easily audited, and easily maintained outside of the supporting Java code. A framework is established, allowing the subject matter experts and data processing support personnel to simulate rule execution in a test environment.

 
 
 
    Copyright © 2000-2017 - EBDX.COM, LLC - All rights reserved.