Objectives

The analysis assists claims administrators in detecting possible fraudulent warranty claims by using data mining and predictive modeling tools. Results summarized in a report can be used by investigators in their case work. The following chart details the relative importance between three variables that could be used in detecting auto warranty claim fraud.

Analysis and Financial Data

The following indicators and tests are applied to warranty claim data:

• Attributes associated with the submitter of the claim or the object subject to the claim (i.e., vehicle’s age, group – make, model – year, part).
• Claims identified as being from phantom customers or for bogus parts.
• Claims submitted by the same customer who received dozens or even hundreds of warranty replacements.
• Claims submitted from many customers all leading to the same address.
• Claim volumes that don’t correspond to quantities of the product actually manufactured or sold in specific geographic areas.
• Claims indicating double billing on the replaced part.
• Claims indicating gouging or unnecessary repairs or parts replacement.
• Claims indicating a possible collusion between the adjuster and the dealer or repair shop.

Benefits

The benefits that can be realized from utilizing these models:

• Reduce the substantial sum of money lost to fraudulent warranty claims by assisting claim administrators in identify participates in the fraud and recover money lost (claim volume may actually decreases as a result of these investigative activities).
• Expand fraud discovery by using CRS models that mine large quantities of claim data for anomalies and fraudulent patterns.