Published on 30 Sep 2009
"Utilizing BRMS, we’ve shortened the inspection processing time for 10,000 cases from two weeks to one day, and field workers can now apply new rules in real time. The more I learn about the many features of IBM WebSphere ILOG BRMS, the more impressed I am with this advanced technology" - Asian Life Insurance Provider
Major Asian Life Insurance Company
Korea - Republic of
Business Process Management (BPM), Service Oriented Architecture
This leading life insurance provider has been a frontrunner of Asia’s insurance industry throughout its history.
Insurance sales are rising fast across Asia Pacific. But along with this rise has come another increase: insurance fraud. This insurance provider needed a new automated insurance fraud detection system (IFDS) to address this growing problem.
The company introduced a new IFDS based on IBM® WebSphere® ILOG JRules, the market-leading business rule management system (BRMS). The system takes knowledge gained from insurance fraud management work and applies it automatically using JRules to prevent false claims from going through. It automates more than 60 percent of the work, while the last system only did about 40 percent, and it has shortened the inspection time for processing 10,000 claims from two weeks to just one day.
Speed-to-market gains of over 50 percent Reduced fraud costs by millions annually Increased productivity by 60 percent
The company offers products and services to customers in the areas of insurance, loans and investment, including pension and savings plans. The company is one of Asia’s largest insurers and a member of the Million Dollar Round Table, an association of insurance planners. It has also been ranked the 18th largest insurer in the world by Fortune magazine.
In South Korea, people are increasingly subscribing to a variety of insurance programs. Not surprisingly, the incidence of insurance fraud has also increased. According to the Korean Financial Supervisory Service, the number of insurance fraud cases in 2006 was 46.4 percent higher than 2005, and the overall trend for financial fraud crimes, such as insurance fraud, lost card fraud and tax evasion, was becoming more diversified and complex. The insurance company, with the equivalent of U.S.$69 billion in reference assets and U.S.$23 billion in sales, and more than 10 million policyholders, needed a new automated IFDS to address this growing problem. The company’s old system was largely a manual process in which the accident processing managers sorted out insurance fraud suspects at their discretion, which increased the probability of human error.
IBM ILOG’s BRMS helps the company scientifically and systematically search and analyze based on 800 different factors, including tracking the history of insurance fraud detections, the number of insurance accident cases and the number of insurance agreements. Fraudulent claims are separated for further inspection by field workers.
New insurance fraud detection methods are entered into the central rule repository as rules and ultimately allow the officers in charge to apply the new rules through the general examination process. With the systematic and fast examination process, the company expects to be able to improve the utilization of its insurance fraud experts. Further, a business user, instead of IT personnel, can now change and add rules themselves in a single day, according to changing inspection requirements, compared to more than two weeks using a manual hard-coding process before.
The BRMS part of the IFDS was designed and developed in six months by KSTEC, IBM’s distributor in South Korea. Written in Java™, the system runs on HP-UX-based PCs, and uses JRules in conjunction with the SAS Statistics Data Mining system, HandySoft BPM Suite and TiMax WAS. The company plans to expand the inspection process to cover compound factors such as family, company and other groups.
The IFDS applies policies expressed in rules using JRules to filter out likely cases of fraud. It helps the insurance provider:
Separate real claims from false ones more quickly
Process real claims in a timely manner
Implement new filtering rules in a day
Capture conditions for detecting fraud in detail
Follow up cases with new rules based on findings from investigations into false claims
Automate more than 60 percent of its claims processing, instead of just 40 percent
Products and services used
• IBM® WebSphere® ILOG JRules
IBM products and services that were used in this case study.
WebSphere ILOG JRules
IBM, the IBM logo, ibm.com and WebSphere are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at ibm.com/legal/copytrade.shtml Other product, company or service names may be trademarks or service marks of others. This case study is an example of how one customer uses IBM products. There is no guarantee of comparable results. References in this publication to IBM products and services do not imply that IBM intends to make them available in all countries in which IBM operates.