Dynamic Workflow Customization via Cognitive Digital Workers

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional business workflows are rigid and static, leading to inefficiencies and inaccuracies, especially in insurance claim processes, where traditional systems lack intelligence and cognitive computing to dynamically manage processes, resulting in repetitive and inaccurate conversations with customers.

Innovation Solution

A cognitive computing-based system that employs digital workers to dynamically customize workflows by selecting the most appropriate workers based on claimant inputs, such as claim history, fraud scores, and external conditions, allowing for early execution of critical steps and continuous learning to refine the workflow, thereby reducing the number of steps required and improving accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional rigid workflows are used, then system simplicity is maintained, but workflow efficiency and adaptability deteriorate

Engineering Contradiction:

Improveworkflow efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent implements dynamic workflow customization by selecting different digital workers based on claim characteristics (fraud score, claim type, amount). The workflow transitions from static to dynamic by using cognitive computing to determine the optimal worker set for each claim, allowing the system to adapt to varying conditions while maintaining high efficiency.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system segments the workflow into distinct steps handled by different digital workers (e.g., fraud detection workers, adjustment workers, customer notification workers). This segmentation allows each worker to specialize in specific tasks, improving overall efficiency while the cognitive computing layer coordinates them dynamically based on claim needs.

Inventive Principle:

Principle #1Segmentation

2Adaptability or versatility

If static workflows are used, then implementation simplicity is maintained, but adaptability to different claim types deteriorates

Engineering Contradiction:

Improveworkflow adaptabilityVSAvoidimplementation simplicity

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The patent creates a universal workflow framework that can handle multiple claim types through a single cognitive computing system. The system uses intent classification and digital worker selection to adapt to different claim scenarios (property damage, bodily injury, fraud) without requiring separate specialized systems for each type, thus achieving versatility while maintaining implementation simplicity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system changes workflow parameters dynamically based on claim characteristics such as fraud score thresholds, claim type, and amount. By adjusting these parameters, the same base workflow can adapt to different claim scenarios, enabling versatility without complex reconfiguration of the underlying system architecture.

Inventive Principle:

Principle #35Parameter changes

3Loss of time

If manual claim processing is used, then flexibility in handling complex cases is maintained, but processing time and cost increase

Engineering Contradiction:

Improveprocessing timeVSAvoidautomation level

Core Design Contradiction:

Loss of timeVSExtent of automation

Solution Approach 1:

The patent implements self-service through digital workers that autonomously handle claim processing tasks. The cognitive computing system automatically selects appropriate workers, executes them, and coordinates the workflow without human intervention for routine claims. This self-service approach reduces processing time while the system maintains flexibility through adaptive worker selection based on claim complexity.

Inventive Principle:

Principle #25Self-service

4Productivity

If repetitive manual tasks are performed, then thoroughness in processing is maintained, but productivity and error rate deteriorate

Engineering Contradiction:

Improveprocessing speedVSAvoidprocessing accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent replaces manual mechanical processing with digital workers that use cognitive computing capabilities. Instead of human employees performing repetitive tasks, the system uses AI-powered digital workers that can process claims faster and with consistent accuracy. The cognitive computing layer ensures reliable worker selection and coordination, maintaining processing accuracy while dramatically improving speed.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11663552B2Dynamically customizing a workflow separate from domain logic

Publication Date: 2023.05.30 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

An approach is provided for improving a workflow by using cognitive computing. Inputs about a claim of a claimant are received. The inputs include a history of claims by the claimant, an expected cost of processing the claim, a length of time the claimant has had insurance coverage, governmental requirements associated with the claim, a fraud score associated with the claim, and a ranking and a clustering of previously processed claims. Based on the inputs and using a cognitive computing system, a set of digital workers is selected. The set of digital workers complete step(s) in a current context of a processing of the claim in a workflow. Responsive to completing the step(s), a total number of steps required in the workflow is reduced.