Back to results
US 11,080,791 B2Granted
Segmentation and balancing system
AssigneeHartford Fire Insurance Company
InventorsPaul J. Edwards, Keren Shemesh
Life of this patent
Estimating the term…
Abstract
According to some embodiments, data may be received indicative of a plurality of insurance claims along with an indication of an appropriate claim segment classification for each insurance claim. A first claim handler may then be automatically selected for a first insurance claim based at least in part on: (i) a first segment classification associated with the first insurance claim, (ii) numbers of other insurance claims currently assigned to claim handlers, and (iii) load factors associated with claim handlers. An indication of the selected first claim handler may then be transmitted.
Claims18
A computerized system, comprising:
- a data storage device storing data indicative of a plurality of insurance claims submitted in connection with insurance policies, the data comprising a plurality of data elements for each of the insurance claims;
- a segmentation platform processor configured for:
- automatically, for a first insurance claim, responsive to the first insurance claim being at least one of a new one of the insurance claims and one of the insurance claims previously segmented having a change in at least one of a subset of the plurality of data elements relating to the first insurance claim:
- determining, based on the data stored by the data storage device associated with the first insurance claim, in accordance with first segmentation logic, a first segment classification for the first insurance claim, the first segment classification being one of a plurality of sequential from least complex to most complex, ranked, complexity segment classifications, wherein at least the first segmentation logic comprises a rule mapping at least one parameter description value to a minimum complexity value of the plurality of sequential segment classifications;
- determining, based on the data associated with the first insurance claim, in accordance with second segmentation logic different from the first segmentation logic, a second segment classification for the first insurance claim, the second segment classification being another one of the plurality of sequential complexity segment classifications,
- comparing the first segmentation classification determined based upon the first segmentation logic and the second segmentation classification determined based upon the second segmentation logic;
- and
- transmitting an indication of the higher of the first and second segment classifications, determined based upon the comparing, to a load balancing and assignment platform processor;
- and
- a load balancing and assignment platform processor in communication with the segmentation platform processor, configured for:
- selecting a first claim handler for the first insurance claim based at least in part on the higher of the first and second sequential segment classifications determined by the segmentation platform processor associated with the first insurance claim, and one or more of (i) numbers of other insurance claims currently assigned to claim handlers, (ii) load factors associated with claim handlers, and (iii) expertise attributes associated with the claim handlers;
- the selecting being performed at least in part by a predictive model trained with historical insurance claim information, the predictive model including at least one of:
- (i) a neural network, and (ii) a support vector machine;
- and
- transmitting an indication of the selected first claim handler.
12 more elements
A computer-implemented method, comprising:
- automatically, by a segmentation platform processor, for a first insurance claim, the first insurance claim being one of a plurality of insurance claims having data relating thereto stored in a data storage device, the data comprising a plurality of data elements for each of the insurance claims, responsive to the first insurance claim being at least one of a new one of the insurance claims and one of the insurance claims previously segmented having a change in at least one of a subset of the plurality of data elements relating to the first insurance claim:
- determining, based on the data stored by the data storage device associated with the first insurance claim, in accordance with first segmentation logic, a first segment classification for the first insurance claim, the first segment classification being one of a plurality of sequential from least complex to most complex, ranked, complexity segment classifications, wherein at least the first segmentation logic comprises a rule mapping at least one parameter description value to a minimum complexity value of the plurality of sequential segment classifications;
- determining, based on the data associated with the first insurance claim, in accordance with second segmentation logic different from the first segmentation logic, a second segment classification for the first insurance claim, the second segment classification being another one of the plurality of sequential complexity segment classifications,
- comparing the first segmentation classification determined based upon the first segmentation logic and the second segmentation classification determined based upon the second segmentation logic;
- and
- transmitting an indication of the higher of the first and second segment classifications, determined based upon the comparing, to a load balancing and assignment platform processor;
- responsive to receipt of the transmitted indication of the higher of the first and second segment classifications, by a load balancing and assignment platform processor in communication with the segmentation platform processor:
- selecting a first claim handler for the first insurance claim based at least in part on the higher of the first and second sequential segment classifications determined by the segmentation platform processor associated with the first insurance claim, and one or more of (i) numbers of other insurance claims currently assigned to claim handlers, (ii) load factors associated with claim handlers, and (iii) expertise attributes associated with the claim handlers;
- the selecting being performed at least in part by a predictive model trained with historical insurance claim information, the predictive model including at least one of:
- (i) a neural network, and (ii) a support vector machine;
- and
- transmitting an indication of the selected first claim handler.
9 more elements
Dependencies are inferred from the claim text.