AI Copilot Task Planning for Legacy System Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional copilot services struggle to efficiently utilize internal information and legacy systems within specific organizations like enterprises, making it difficult to adapt to diverse environments and provide effective task support.
Innovation Solution
An AI-based copilot service that calculates a task plan comprising unit tasks, interworks with sub-systems including internal databases and legacy systems, and uses AI models to generate responses tailored to user requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional copilot services use pre-learned general knowledge or reflected information from user requests, then the service can operate without complex system integration, but the service cannot efficiently utilize internal information and legacy systems of specific organizations
Solution Approach 1:
The system segments the complex task of utilizing organizational resources into distinct unit tasks (e.g., searching internal information, accessing legacy systems, generating responses). Each unit task is handled by specialized sub-systems, making the overall complex system manageable and adaptable to different organizational environments through modular configuration.
Solution Approach 2:
The copilot service is designed with universal interfaces that can connect to various types of sub-systems (databases, legacy systems, APIs) through standardized protocols. This multi-functionality allows the same copilot platform to adapt to different organizational environments without requiring complete system redesign, thus improving adaptability while managing complexity through standardization.
2Reliability
If the copilot service integrates with multiple sub-systems to access internal information, then the service can provide organization-specific answers, but the configuration and management of diverse database structures and legacy systems becomes difficult
Solution Approach 1:
The system introduces intermediary layers (abstraction layers, adapters, or middleware) between the copilot service and diverse sub-systems. These intermediaries translate between the standardized copilot interface and the varied internal structures of different databases and legacy systems, enabling accurate organization-specific answers while simplifying configuration and management through a unified interface.
Solution Approach 2:
The system allows configuration of connection parameters and interaction modes for different sub-systems through parameterization. Instead of hardcoding connections to specific database structures, the system uses configurable parameters that can be adjusted to match different organizational environments, maintaining reliability of organization-specific answers while improving ease of operation through flexible parameter setup.
3Adaptability or versatility
If the copilot service uses a fixed configuration for internal information access, then the system structure is simple, but the service cannot flexibly reflect various environments of specific organizations
Solution Approach 1:
The system transitions from fixed configuration to dynamic configuration, where the copilot service can adapt its sub-system connections and interaction parameters based on the specific organizational environment. This dynamic capability allows flexible reflection of various organizational structures while managing complexity through automated configuration and standardized interfaces that reduce manual setup requirements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A processor-implemented method including calculating a task plan, the task plan including one or more unit tasks, based on a user request, performing the task plan in association with one or more sub systems corresponding to the one or more unit tasks, and providing an answer to the user request, based on a result of performing the task plan.