Hybrid AI Support System for Reducing Labor Intensity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current web applications for user inquiry handling are resource-intensive and often difficult to navigate, with live support systems being labor-intensive and lacking assurance in providing solutions, while chat-bot support struggles with users describing their inquiries accurately and relying on scripted solutions.

Innovation Solution

A hybrid system and platform that combines automated and live support by processing user inquiries, streaming the application or web browser state, and using natural language processing to determine if an automated response is available, providing either automated or live responses based on availability, and allowing for the automation of recent inquiries.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If live support is provided to users, then user service quality is improved, but labor intensity increases

Engineering Contradiction:
Improveuser service qualityVSAvoidlabor intensity
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The support system is segmented into automated response generation and live human review components. The automated system handles routine inquiries and generates initial responses, while human agents only intervene for complex or unsatisfied cases, thereby maintaining service quality while reducing overall labor intensity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

An automated AI system acts as an intermediary between users and human support agents. The AI processes inquiries, generates draft responses, and filters cases before reaching human agents, reducing the burden on human workers while maintaining reliable service delivery

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If chat-bot support is used, then labor intensity is reduced, but ability to handle complex inquiries deteriorates

Engineering Contradiction:
Improvelabor intensityVSAvoidability to handle complex inquiries
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system merges automated chat-bot capabilities with live human support into a hybrid model. The automated system handles initial processing and routine cases, while seamlessly transitioning to human agents for complex inquiries, combining the efficiency of automation with the problem-solving ability of humans

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The automated system performs preliminary actions by processing inquiries, generating draft responses, and preparing case information before human agents become involved. This preliminary work reduces the complexity of tasks human agents must handle while maintaining the ability to resolve complex issues

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If web applications are made more comprehensive, then functionality is improved, but ease of navigation deteriorates

Engineering Contradiction:
ImprovefunctionalityVSAvoidease of navigation
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system incorporates self-service capabilities that automatically guide users through the appropriate support processes based on their inquiries. The automated system navigates users to relevant resources and support options without requiring them to manually navigate complex application structures

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS20230409616A1Hybrid guided communication session with artificial intelligence
Publication Date: 2023.12.21 US BANK NATIONAL ASSOCIATION
  • US20230409616A1 patent drawing
  • US20230409616A1 patent drawing
  • US20230409616A1 patent drawing

AI summary

Systems, methods, and computer-readable media are provided for hybrid guidance of a communication session. In one aspect, a system includes one or more processors configured to execute computer-readable instructions to receive an inquiry from a user terminal; stream the inquiry to one or more support terminals configured to provide live support for the inquiry; determine if an automated response to the inquiry is available; and provide one of the automated response or a live response to the inquiry based on whether the automated response to the inquiry is available or not.